View
216
Download
1
Category
Preview:
Citation preview
Understanding the Function of PGC-1α Isoforms in
β-cell Survival and Diabetes
Sarah Sczelecki
Faculty of Medicine
Division of Experimental Medicine
McGill University
Montréal, Quebec, Canada
Supervisor: Dr. Jennifer L. Estall
Co-Supervisor: Dr. Woong-Kyung Suh
Unité de mécanismes moléculaires du diabète
Institut de recherches cliniques de Montréal (IRCM)
A thesis submitted to McGill University in partial fulfillment of the requirements of the
Master of Science degree
© Sarah Sczelecki 2013
1
Table of Contents
Table of Contents 1
Abstract 3
Résumé 4
Acknowledgements 6
List of Abbreviations 7
SECTION I: INTRODUCTION 12
The Pancreas and β-cells 13
Diabetes: A Global Pandemic 14
Type 1 Diabetes 15
Type 2 Diabetes 17
β-cell Death and Apoptosis 19
Glucagon-like Peptide 1 (GLP-1) and Diabetes 20
Mouse Models of Diabetes 22
Peroxisome proliferator activated receptor gamma co-activator 1-alpha (PGC-1α) 24
PGC-1α and β-cell Biology 29
PGC-1α Isoforms 31
PGC-1α and Disease 35
Objectives 37
SECTION II: MATERIALS AND METHODS 38
Cell Culture 39
Islet Isolation 39
RNA Extraction and cDNA Synthesis 40
Quantitative Real-Time PCR (qPCR) 40
Protein Analysis 41
Genotyping 41
Induction of PGC-1α Isoforms 42
Cleaved Caspase-3 Assay 43
PGC-1α Knock-out and Streptozotocin in vivo Experiment 43
Enzyme-linked Immunosorbent Assay (ELISA) 44
Immunohistochemistry 45
Cleaved Caspase-3 Quantification 45
Statistical Analysis 46
SECTION III: RESULTS 47
PGC-1α isoforms are differentially expressed in a rodent β-cell line and
primary islets 48
Cytokines modulate the expression of PGC-1α1 and PGC-1α4 50
PGC-1α4 prevents the cleavage of caspase-3 in response to cytokines 52
PGC-1α1 and PGC-1α4 expression are regulated by GLP-1 54
2
Knock-down of PGC-1α isoforms, prevents streptozotocin induced
hyperglycemia in vivo 55
SECTION IV: DISCUSSION 59
Summary of Results 60
Regulation and Function of PGC-1α Isoforms in β-cells 61
Isoforms and the Pathogenesis of Diabetes 66
Differential Role of PGC-1α1 and PGC-1α4 in β-cells 70
PGC-1α4 versus NT-PGC-1α 71
Perspectives and Conclusions 73
SECTION V: FIGURES 77
Figure 1. Summary of extrinsic and intrinsic pathways of apoptosis 78
Figure 2. Anti-apoptotic actions of GLP-1 79
Figure 3. Functional domains of PGC-1α 80
Figure 4. Summary of PGC-1α isoform transcripts 81
Figure 5. Nucleotide and protein sequence similarities between
NT-PGC-1α and PGC-1α4 82
Figure 6. Genotyping of mouse tail crude DNA 83
Figure 7. Endogenous expression of isoforms in a β-cell line and primary islets 84
Figure 8. Incubation with a cytokine cocktail regulates the expression of
PGC-1α1 and PGC-1α4 in a time-dependent manner. 85
Figure 9. Adenoviral over-expression of PGC-1α1 and PGC-1α4 reduces
cleavage of caspase-3 86
Figure 10. Treatment of INS-1 cells with Exendin-4 induces the expression of
PGC-1α1 and PGC-1α4 87
Figure 11. Experimental plan of PGC-1α KO with low-dose STZ injections 88
Figure 12. Knockout of PGC-1α and its isoforms in vivo 89
Figure 13. Cleaved caspase-3 expression in islets 1 day post-STZ injections
of WT vs. KO mice 90
Figure 14. Cleaved caspase-3 expression in islets 1 day post-STZ injections
of WT vs MIP-Cre control mice 91
SECTION VI: TABLES 92
Table 1: List of PCR Primers 93
Table 2: List of qPCR Primers 93
References 94
3
Abstract
Peroxisome proliferator-activated receptor gamma (PPARγ) co-activator 1 alpha
(PGC-1α) is a transcriptional co-activator responsible for mitochondrial biogenesis and oxidative
metabolism. Many isoforms of PGC-1α have been described in the literature, most of which are
shown to function similarly to the canonical PGC-1α protein. Recently, however, a novel
isoform of PGC-1α was identified, PGC-1α4. It was shown to have a different, yet
complementary function to canonical PGC-1α (PGC-1α1) in muscle. It is also expressed in other
metabolically active tissues; however, it is unknown whether it has additional distinct
tissue-specific functions. Furthermore, PGC-1α plays an important role in controlling
metabolism in pancreatic β-cells and expression of the co-activator is decreased in diabetic islets;
however, the role of PGC-1α isoforms in diabetes is unknown.
Our objective is to determine whether PGC-1α4 has a unique function in β-cells and
whether it plays a role in the pathogenesis of diabetes. We show that stimulation with forskolin,
exendin-4 and a cytokine cocktail of TNF , IL-1 and IFN , induced specific PGC-1 isoforms
in -cells. Following over-expression of these isoforms in INS-1 cells, PGC-1α4 prevented the
cleavage of caspase-3 in response to cytokines, suggesting that the novel isoform is uniquely
anti-apoptotic. To assess whether PGC-1α isoforms play a role in β-cell survival in vivo, mice
with a β-cell specific PGC-1α knockout of all isoforms were subjected to low-dose
streptozotocin (STZ) treatment to induce β-cell apoptosis. Unexpectedly, knockout mice were
protected from STZ induced hyperglycemia. However, there was no difference in percentage of
cleaved caspase-3 positive cells in control versus knockout mice, suggesting no difference in
apoptosis. Therefore, PGC-1α4 could be a novel factor important for β-cell survival and over-
expression of this unique isoform may protect against the pathogenesis of diabetes.
4
Résumé
Peroxisome proliferator-activated receptor gamma (PPARγ) co-activator 1 alpha
(PGC-1α) est un co-activateur transcriptionnel responsable de la biogenèse mitochondriale et du
métabolisme oxydatif. De nombreux isoformes de PGC-1α ont été décrits dans la littérature, dont
la plupart ont été démontrés comme fonctionnant de manière similaire à la protéine PGC-1α
canonique. Récemment, cependant, un nouvel isoforme de PGC-1α a été identifié, PGC-1α4,
possédant une fonction différente mais complémentaire de la protéine canonique (PGC-1α1)
dans le muscle. PGC-1α4 est également exprimé dans d'autres tissus métaboliquement actifs,
mais il n’est pas connu s’il démontre de nouvelles fonctions dépendantes du tissu. PGC-1α joue
un rôle important dans la regulation métabolique des cellules β du pancreas; de plus l’expression
du co-activateur est dérégulée dans les îlots de Langerhans de patients diabétiques.
Notre objectif est de déterminer si PGC-1α4 a une fonction unique dans les cellules β du
pancréas et s’il joue un rôle dans la pathogenèse du diabète. Nous montrons que la stimulation
avec la forskoline, l'exendine-4 et un cocktail de cytokines, TNF , IL-1 et IFN , induit les
isoformes de PGC-1α dans les cellules β. Suite à la surexpression des isoformes dans les cellules
INS-1, PGC-1α4 empêche le clivage de la caspase-3 en réponse à des cytokines, suggérant
qu’uniquement le nouvel isoforme est anti-apoptotique. Pour déterminer si les isoformes de
PGC-1α jouent un rôle dans la survie des cellules β in vivo, les souris déficientes de tous les
isoforms de PGC-1α, spécifiquequement dans les cellules β, ont été soumises à une faible dose
de streptozotocine (STZ) provoquant l'apoptose des cellules β. De façon inattendue, les souris
déficientes en PGC-1α ont été protégées contre l’hyperglycémie induite par la STZ. Cependant,
il n'y avait pas de différence du pourcentage de cellules en apoptose chez les souris témoins par
rapport aux souris déficientes. Donc, PGC-1α4 pourrait être un nouveau facteur important pour
5
la survie des cellules β du pancreas et la surexpression de cette isoform unique, peut protéger
contre la pathogenèse du diabète.
6
Acknowledgements
I would like to express my sincere gratitude to Dr. Jennifer Estall for her unwavering
support and guidance in these past two years – she has shaped me into the scientist that I am
today. I would also like to thank my co-supervisor Dr. Woong-Kyung Suh for his insightful input
into my project.
I am grateful to the Estall lab members for assisting me with my project and making the
lab a pleasant place to come to everyday. Daniel – I don’t know what I would have done without
your ninja gavaging skills and flawless islet isolating techniques. Aurèle – thank you for your
endless support in the lab (saving the day with the MIP-Cre experiment) and making every day
fun and a little bizarre. Also, thank you for reading my thesis and fixing my horrible French.
Adam – thank you for cutting the blocks. Alex – words cannot even express how grateful I am to
have started in the lab with you there. Not only did you give me confidence in my abilities, you
were there for some good chats and a lot of Adele. I am lucky to have you as a friend.
I would also like to thank Dominique Lauzier for her assistance in histology; Dominic
Filion for help with microscopy and; Marie-Claude Lavallée, Jade Dussureault, Caroline Dubé
and Ève-Lyne Thivierge for all their assistance with the mice.
To the Primers – my time here wouldn’t have been the same without you guys. Go
Primers! Loups garous!
Last, but certainly not least, to my family for their love and support.
7
List of Abbreviations
AAV8 Adeno-associated virus 8
Ad-siPGC-1α Adenoviral vector short interfering RNA against PGC-1α
ATMs Adipose tissue macrophages
ATP5B ATP synthase subunit β
ASO Anti-sense oligonucleotides
BAT Brown adipose tissue
bp base pairs
BSA Bovine serum albumin
BMI Body Mass Index
cAMP Cyclic adenosine monophosphate (AMP)
CBP CREB binding protein
CMV Cytomegalovirus
COXII Cytochrome c oxidase subunit II
COXIV Cytochrome c oxidase subunit IV
COXVb Cytochrome c oxidase subunit V b
CPT-1 Carnitine palmitoyl transferase I
CREB cAMP response element-binding protein
CTLA-4 Cytotoxic T-lymphocyte Antigen 4
CVD Cardiovascular disease
CytoC Cytochrome C
DOX Doxycycline
DPP-4 Dipeptidyl peptidase-4
ERRα Estrogen-related recptor α
8
ETC Electron transport chain
Ex-4 Exendin-4
FFA Free fatty acids
FK Forskolin
FOXO1 Forkhead box protein O1
G6pc Glucose-6-phosphatase
GAD65 Glutamic acid decarboxylase 65
GFP Green fluorescent protein
GLP-1 Glucagon-like peptide 1
GLP-1R Glucagon-like peptide 1 receptor
Glut2 Glucose transporter type 2
Glut4 Glucose transporter type 4
GSIS Glucose stimulated insulin secretion
HAT Histone acetyltransferase
HBSS Hanks balanced salt solution
HLA Human leukocyte antigen
HNF4α Hepatocyte nuclear factor 4 alpha
HPRT Hypoxanthine-guanine phosphoribosyltransferase
I-A2 Insulinoma-associated antigen-2
IAP Inhibitors of apoptosis
IDDM1 Insulin-dependent diabetes mellitus locus 1
IGF1 Insulin-like growth factor 1
IHC Immunohistochemistry
9
IL-1β Interleukin-1beta
IL2RA Interleukin 2 receptor A
IL6 Interleukin 6
IFNγ Interferon gamma
iNOS Nitric oxide synthase
INS-1 Rat insulinoma β-cell line
IP Immunoprecipitation
ip Intraperitoneal
kDa Kilo Daltons
KO Knock-out
LCAD Long chain acyl-coenzyme A dehydrogenase
MCAD Medium chain acyl-coenzyme A dehydrogenase
MCP-1 Monocyte chemoattractant protein 1
MEF2 Myocyte enhancer factor-2
MIP Mouse insulin promoter
MSN Myostatin 1
mtTFA Mitochondrial transcription factor A
NOD Non-obese diabetic
NRF1 Nuclear respiratory factor-1
NRF2 Nuclear respiratory factor-2
ns Not significant
NT-PGC-1α N-truncated PGC-1α
OGTT Oral glucose tolerance test
10
Pck1 Phosphoenol pyruvate carboxykinase 1
PCR Polymerase chain reaction
PEPCK Phosphoenol pyruvate carboxykinase
PGC-1α Peroxisome proliferator activated receptor gamma co-activator 1-alpha
PGC-1α1 PGC-1α isoform 1, canonical transcript
PGC-1α2 PGC-1α isoform 2
PGC-1α3 PGC-1α isoform 3
PGC-1α4 PGC-1α isoform 4
PGC-1α-b PGC-1α isoform b
PGC-1α-c PGC-1α isoform c
PGC-1β Peroxisome proliferator activated receptor gamma co-activator 1-beta
PPARγ Peroxisome proliferator activated receptor gamma
PPARα Peroxisome proliferator activated receptor alpha
PRC PGC-related coactivator
PTPN22 Protein tyrosine phosphatase, non-receptor type 22
qPCR Quantitative real-time polymerase chain reaction
ROS Reactive oxygen species
siRNA Short interfering RNA
SRC-1 Steroid receptor coactivator-1
STZ Streptozotocin
T1D Type 1 Diabetes
T2D Type 2 Diabetes
TetO Tetracycline-dependent promoter
11
Tfam Mitochondrial transcription factor A
tTA Tetracycline transactivator
TNFα Tumor necrosis factor alpha
TUNEL Terminal deoxynucleotidyl transferase dUTP nick end labeling
UCP-1 Uncoupling protein-1
UCP-2 Uncoupling protein-2
WAT White adipose tissue
WT Wild-type
ZNT8 Zinc transporter 8
12
SECTION I:
INTRODUCTION
13
The Pancreas and β-cells
The pancreas is a mixed glandular organ consisting of both exocrine and endocrine cells.
Exocrine cells are organized into pancreatic acini and secrete digestive enzymes, whereas the
endocrine pancreas is organized into Islets of Langerhans, which secrete hormones
(Collombat et al., 2010). These hormones include insulin, glucagon, somatostatin, pancreatic
polypeptide and ghrelin, from the β-, α-, δ-, γ- and ε-cells, respectively (Collombat et al., 2010;
Elayat et al., 1995). In humans, the different cells of the islet are interspersed throughout the
whole structure, whereas in mice the α-cells (and δ-cells) surround the periphery and the β-cells
are found primarily in the center of the islet (Kim et al., 2009). β-cells make up approximately
80% of the islet, whereas the other cell types comprise the other 20%. The main function of
β-cells is to store and release insulin in response to glucose. Insulin is required in the blood to
maintain glucose homeostasis and not surprisingly, β-cell dysfunction or death results in the
development of diabetes. The main function of α-cells is to secret glucagon in times of
starvation, to increase circulating glucose by initiating gluconeogenesis in the liver. Impaired
glucagon secretion is also observed in type 2 diabetics (Del Prato and Marchetti, 2004). In type 2
diabetics, fasting plasma glucagon concentrations are elevated and there is impaired glucagon
suppression after intake of a meal (Del Prato and Marchetti, 2004). These events contribute to
hyperglycemia and eventual β-cell dysfunction, by maintaining hepatic glucose output when in
the fed state. However, β-cell dysfunction and death remains the main cause of diabetes
development and target for therapies.
14
Diabetes: A Global Pandemic
Both type 1 (T1D) and type 2 (T2D) diabetes mellitus are becoming a global concern.
90% of diabetics have T2D, which correlates with increased incidence of obesity, reduced
physical behaviour and dietary changes. The remaining 10% of diabetics are classified as type 1.
Currently, it is predicted that 347 million people worldwide are affected by diabetes (Danaei et
al., 2011). By 2030, diabetes incidence is projected to increase by 50% and become the 7th
leading contributor to increased mortality worldwide (Mathers and Loncar, 2006). Even though
the prevalence of diabetes is higher in developed countries, the prevalence is rapidly increasing
in developing countries (Forbes and Cooper, 2013).
Mortality associated with either type of diabetes is due to complications of the disease
including cardiovascular disease and diabetic nephropathy, which are the leading causes of death
for diabetic patients (Morrish et al., 2001). Globally, a total of 465 billion US dollars (USD) was
spent on treating diabetes and its complications in 2011 and by 2030, this expense is projected to
increase to 654 billion by 2030 (IDF, 2013).
T2D can be treated by lifestyle intervention if detected early on in disease progression,
drastically improving symptoms and reducing the risk of developing complications. However,
later stages of T2D cannot simply be treated with lifestyle changes and requires pharmacological
intervention (Prentki and Nolan, 2006). Moreover, unlike T2D, T1D cannot be treated simply
with lifestyle changes, but requires use of exogenous insulin and drugs to maintain euglycemia
and to reduce the risk of diabetic complications. Overall, there is a need to further understand the
underlying molecular basis of either disease to improve treatment efficacy to prevent diabetes
associated mortality (van Belle et al., 2011).
15
Type 1 Diabetes
T1D is a multi-factorial disease, occurring due to a combination of environmental and
polygenic factors (Forbes and Cooper, 2013) and described as insulin-dependent diabetes. It
develops following immune mediated attack of the insulin producing β-cells in the Islets of
Langerhans resulting in cell death (Padgett et al., 2013). The use of exogenous insulin controls
glycemia; however, chronically elevated levels of glucose and insulin cause microvascular
insults increasing the risk of cardiovascular disease (Reusch and Wang, 2011). Briefly,
chronically high levels of circulating glucose cause an increase in glucose uptake in endothelial
cells, which are unable to efficiently reduce glucose intake. This increases the amount of electron
donors, such as NADH and FADH2, being fed into the electron transport chain (ETC)
(Brownlee, 2005). This inevitably generates electron radicals, which are able to bind oxygen to
generate superoxide, a reactive oxygen species (ROS), which ultimately leads to endothelial cell
dysfunction and death leading to microvascular disease (Brownlee, 2005).
The cause of type 1 diabetes is largely unknown; however, infectious insults or mutations
in particular genes have been linked to the development of the disease. Many of the genes
associated with the development of T1D are known and most are associated with immunological
function. One class of genes mutated in T1D are the HLA (human leukocyte antigen) genes,
HLA class I and II, located in a region known as the insulin-dependent diabetes mellitus locus
(IDDM1) (van Belle et al., 2011). Other genes associated with the development of diabetes are
(i) the insulin gene (Bell et al., 1984); (ii) PTPN22 (protein tyrosine phosphatase, non-receptor
type 22), a negative regulator of T-cell function; (iii) IL2RA (interleukin 2 receptor A) , a Treg
cell survival signal and; (iv) CTLA-4 (cytotoxic T-lymphocyte Antigen 4), a crucial molecule for
the negative regulation of inflammatory signalling (reviewed in van Belle et al., 2011).
16
Not surprisingly, a majority of the genes associated with T1D relate to dysregulated
inflammatory signalling, including loss of T-cell regulation and Treg cell function. Type 1
diabetes is characterized by immune infiltration of CD4+ and CD8+ T-cells around the islets, as
well as the presence of macrophages, B-cells, NK cells and NKT cells (reviewed in Bending et
al., 2011). Auto-immune attack of β-cells by these cell types is caused by a loss of tolerance to
self-antigens, such as GAD65 (glutamic acid decarboxylase 65) (Karlsen et al., 1992), I-A2
(insulinoma-associated antigen-2), proinsulin, and ZnT8 (Zinc transporter 8) (van Belle et al.,
2011). Presence of immune cells surrounding the islet exposes the β-cells to chronic levels of
pro-inflammatory cytokines, such as TNFα (Tumor necrosis factor alpha), IL-1β (Interleukin 1
beta) and IFNγ (Interferon gamma) (Donath et al., 2003). Interestingly, acute exposure of β-cells
to these cytokines can increase β-cell function; however, chronic exposure to pro-inflammatory
cytokines can lead to β-cell death.
Immune infiltration of the islet leads to β-cells death via various pathways. In a chronic
context, β-cells up-regulate pro-inflammatory pathways, such as NF-κB and JAK/STAT1, which
are detrimental to β-cell health (van Belle et al., 2011). As well, β-cells increase the amount ROS
and decrease pro-survival genes, like Bcl-2. It is thought that pro-inflammatory cytokines can
increase iNOS (Nitric oxide synthase), leading to nitric oxide accumulation and eventually
caspase activation and β-cell death. Another pathway involved in β-cell death is an increase in
signalling of the death receptor, Fas. In a non-disease state, Fas receptor expression is negligible;
however, upon exposure to pro-inflammatory cytokines, Fas expression increases and causes
islets to be more susceptible to death (van Belle et al., 2011). Moreover, cell death promotes
antigen presentation, perpetuating the auto-immune attack of the β-cells and further promoting
their death (van Belle et al., 2011). The culmination of all these events, whether acting
17
simultaneously or individually, results in decreased β-cell mass, reducing circulating insulin and
causing the severe hyperglycemia characteristic of type 1 diabetes.
Type 2 Diabetes
Type 2 diabetes is known as insulin-independent diabetes and is associated with insulin
resistance and hyperglycemia. In early stages, type 2 diabetics exhibit insulin resistance in their
peripheral tissues, but at this stage β-cells are able to compensate and increase secretion of
insulin maintaining euglycemia. This compensation in rodents is due to an increase in β-cell
mass (Steil et al., 2001) and function (Chen et al., 1994). The compensatory increase in β-cells
mass also occurs in type 2 diabetics (Butler et al., 2003); however, the extent to which β-cells in
humans can expand seems limited and the source of the expanded β-cell pool remains unclear
and is highly debated. New β-cells may come from pre-existing β-cells (Dor et al., 2004) or are
derived from the ductal epithelium (Inada et al., 2008; Lee et al., 2010), where β-cell stem-like
cells are thought to reside. Regardless of the origin, this is an important mechanism to
understand and can serve as a potential therapy for T2D, since over time the compensatory β-cell
mass increase appears to be lost and type 2 diabetics begin to become hyperglycemic.
Many mechanisms contribute to the loss of β-cell function in T2D. Since T2D incidence
relates to over-nutrition and inactivity, there is often an increase of both free-fatty acids (FFAs)
and glucose in the blood, which are all processed by the mitochondria. Overuse of the
mitochondria over time, increases ROS, due to more substrates entering the ETC, generating free
electrons and eventually leads to mitochondrial dysfunction (Lowell and Shulman, 2005;
Maestre et al., 2003). β-cells are more susceptible to ROS, as they have comparatively lower
18
amounts of detoxifying enzymes (reviewed in Prentki and Nolan, 2006). This inability to
detoxify free radicals eventually leads to β-cell death via apoptosis (Maestre et al., 2003).
In addition to over-nutrition causing an overload on metabolic machinery, there is also an
inflammatory component that contributes to the progression of the disease. However, these two
processes are difficult to separate. Over-nutrition in its extreme form causes obesity, leading to
increased fat storage in adipocytes and generating larger fat pads. This increase in fat mass also
coincides with an increase in resident macrophages, known as adipose tissue macrophages
(ATMs) (Weisberg et al., 2003; Xu et al., 2003). An increase of circulating FFAs activates the
ATMs via TLR4 receptors (Huang et al., 2012; Nguyen et al., 2007), causing them to secrete a
number of pro-inflammatory cytokines, such as, TNFα, IL-1β and IL-6 (interleukin-6)
(Fain, 2006). High levels of circulating TNFα contributes to insulin resistance and poor glucose
tolerance (Hotamisligil et al., 1993) and inflammatory markers are also predictive for type 2
diabetics. For example, IL1RA is elevated in obese patients before the onset of T2D as a
compensatory mechanism to counteract the increase in IL-1β levels (Meier et al., 2002). An
increase in circulating cytokines provided an environment for chronic exposure of the β-cells to
these harmful stimuli, and similar to T1D, is thought to lead to β-cell death by apoptosis.
In later stages of T2D, there is also hyperglycemia caused by β-cell dysfunction and
death. Similar to T1D, type 2 diabetics are also at risk to develop micro- and macrovascular
complications, such as cardiovascular disease (CVD), retinopathy and renal failure (Murea et al.,
2012). Thus both T1D and T2D, despite having different initial causes, are similar in their
inflammatory aspects causing β-cell dysfunction and eventually death. The eventual outcome of
both diseases is a loss of the insulin producing cells, causing a dysregulation of glucose
homeostasis, which can only be managed by pharmacological intervention or islet transplantation
19
treatments. Therefore, understanding how to limit β-cell death can be applied therapeutically to
both types of diabetes.
β-cell Death by Apoptosis
β-cell death in diabetes is linked to apoptosis. β-cell apoptosis can be mediated through
an extrinsic or intrinsic pathway. Cell death via the extrinsic pathway occurs through Fas and
FasL interaction (Peter and Kramer, 1998). Pro-inflammatory cytokine exposure of islets induces
the expression of Fas on β-cells, making them more susceptible to FasL binding. Additionally,
TNFα binding to its cognate receptor (TNFR) (Locksley et al., 2001) also activates the extrinsic
pathway. These receptors bind their adaptor protein, FADD and TRADD, respectively, which
cleave and activate caspase-8, ultimately leading to apoptosis through activation of caspase-3
(Figure 1) (Locksley et al., 2001; Peter and Kramer, 1998). Caspases, or cysteine-aspartic
proteases, are first inactive pro-caspase proteins that are then cleaved into two pieces, which
need to hetero-dimerize to become active to carry out their functions (Alnemri et al., 1996).
The intrinsic pathway also is an important arm of cell death by apoptosis. This is cell
death based on non-receptor mediated activation of the apoptosis pathway and is initiated by the
mitochondria (Elmore, 2007). Apoptotic cues such as loss of growth signals or exposure to
cytokines causes inner mitochondrial membrane changes, leading to the release of pro-apoptotic
proteins cytochrome-c (CytoC) (Garrido et al., 2006) and Smac/DIABLO (Du et al., 2000) from
the inner membrane space to the cytosol. Once in the cytosol, these proteins activate caspase-9
activity or inhibit IAP (inhibitors of apoptosis), respectively, committing the cell to undergo
apoptosis (Elmore, 2007). The mitochondrial apoptotic events are mediated by Bcl-2 family
20
members, such as Bcl-2 and Bcl-XL which are anti-apoptotic and Bax, Bid, and Bim, which are
pro-apoptotic (Cory and Adams, 2002). Bcl-2 and Bcl-XL inhibit mitochondrial membrane
permeabilization, effectively preventing the release of CytoC. Whereas, Bim, Bax and Bid, can
either inhibit Bcl-2 and Bcl-XL, or promote membrane permeabilization, causing CytoC release
(Cory and Adams, 2002). There is cross-talk between the extrinsic and intrinsic pathways, since
the Fas pathway can cause mitochondrial damage leading to activation of Bid (Figure 1)
(Elmore, 2007).
Both of these pathways converge on executioner caspases, such as caspase-3, -6, -7,
which are ultimately responsible for the changes seen in apoptotic cells, including DNA
fragmentation and membrane blebbing (Elmore, 2007; Cory and Adams, 2002). Caspase-3 can
be activated by the most initiator caspases (caspase-8, -9 or -10) and once activated, leads to
DNA fragmentation and formation of apoptotic bodies, irreversibly committing the cell to
apoptosis (Elmore, 2007).
Glucagon-like Peptide 1 (GLP-1) and Diabetes
The discovery that β-cell death is an integral part of the pathogenesis of diabetes lead to
great interest in finding chemical entities, both endogenous and synthetic, that could prevent
β-cell damage and apoptosis. Glucagon-like peptide-1 (GLP-1) receptor (GLP-1R) agonists are a
class of pharmacological intervention therapies that promote survival of β-cells and are already
being used to treat diabetes. GLP-1 is encoded in the proglucagon gene and is a posttranslational
proteolytic product of the proglucagon precursor. It is secreted from L-cells found in the more
distal regions of the intestine. Primarily, GLP-1 secretion is post-prandial, particularly in
21
response to a meal high in fats and carbohydrates. One of the main functions of the hormone is to
enhance glucose stimulated insulin secretion from β-cells (Baggio and Drucker, 2007). The half-
life of GLP-1 is under 2 minutes due to rapid inactivation by DPP-4 (dipeptidyl peptidase-4)
(Deacon et al., 1995), limiting this use of the native form of the peptide in a clinical setting.
GLP-1 binds to its cognate receptor, GLP-1R, which is a seven pass transmembrane
GPCR receptor. Interestingly, GLP-1R can also bind other agonists such as Exendin-4.
Exendin-4 was originally isolated from the saliva of the Heloderma suspectrum, and is a long-
lasting analog of GLP-1, which can activate mammalian GLP-1R and functions the same as
native GLP-1 (Baggio and Drucker, 2007). For example, agonist binding of GLP-1R on β-cells
activates adenylyl cyclase to stimulate intracellular cyclic AMP (cAMP) production (Figure 2).
Activation of cAMP in β-cells increases insulin secretion and promotes β-cell survival. Cyclic
AMP increases insulin secretion in β-cells by ultimately increasing intracellular calcium
concentrations, promoting insulin release. Moreover, cAMP signalling in the β-cell confers
GLP-1’s cyto-protective effect either through activation of CREB (cAMP response element-
binding protein) leading to Bcl-2 and Bcl-XL expression or suppression of caspase-3 activation,
following Akt-PKB activation (Figure 2) (Baggio and Drucker, 2007; Jhala et al., 2003; Wang
and Brubaker, 2002).
Defects in insulin secretion and β-cell death are hallmarks of the pathophysiology of
diabetes. GLP-1 can potentiate insulin secretion and is cyto-protective, making GLP-1R agonists
effective treatments for type 2 diabetes. In fact, the incretin response is deficient after food intake
in type 2 diabetic patients (Nauck et al., 1993). However, because GLP-1 has a short half-life,
long lasting agonists are used, such as Exendin-4. Indeed, Exendin-4 is more potent than GLP-1
in lowering blood glucose in diabetic animal models (Young et al., 1999), and in diabetic
22
patients (Fineman et al., 2003), using the pharmacological analog, Exenatide. An alternative
drug, Liraglutide, is more stable than Exenatide and just as efficient. Because of its increased
stability, only one injection daily of Liraglutide is needed as compared to two for Exenatide
(Degn et al., 2004). Alternatively, drug therapy to inhibit the action of DPP-4 is also effective to
prevent the rapid degradation of secreted and circulating native GLP-1. Drugs to inhibit DPP-4
action, such as Vildagliptin and Sitagliptin, increase plasma concentrations of incretin hormones
including GLP-1, and enhance GSIS and β-cell function (Baggio and Drucker, 2007). Even
though GLP-1R agonists or DPP-4 inhibitors are being used to treat T2D, their effects may also
be beneficial for type 1 diabetics. Exenatide improves insulin secretion in patients with
transplanted islets; however, there is no evidence to suggest any change in β-cell survival or
proliferation (Ghofaili et al., 2007).
GLP-1R agonists have been therefore shown to improve glucose homeostasis in diabetic
patients mostly by potentiating insulin secretion. There is limited evidence to suggest that
GLP-1R agonists and DPP-4 drugs can impact β-cell growth and survival in patients. Therefore,
understanding the molecular mechanisms responsible for diabetes and potentiating β-cell growth
and survival pharmacologically could enhance the efficacy of this therapeutic option (Lovshin
and Drucker, 2009).
Mouse Models of Diabetes
To better understand the molecular mechanisms of human diabetes and pharmacological
interventions, mouse models of diabetes are often utilized. Common mouse models to mimic
type 1 diabetes mouse models are streptozotocin (STZ) treated mice or non-obese diabetic
23
(NOD) mice. Popular models of type 2 diabetes are the ob/ob, db/db genetic models, or mice fed
a high-fat diet to cause diet induced obesity and diabetes (Sakata et al., 2012). Type 2 rodent
diabetes models are more related to an increased incidence of insulin resistance. Moreover,
initially β-cells in type 2 mouse models are able to proliferate to compensate for the insulin
resistance occurring in the periphery (Prentki and Nolan, 2006) so β-cell death is not immediate.
Therefore, to study β-cell death early on, type 1 diabetes models may be more relevant.
STZ is a chemical toxin similar in structure to glucose and is therefore able to enter cells
using the Glut2 glucose transporters (Schnedl et al., 1994; Szkudelski, 2001). In the islet, only
β-cells express Glut2, thus STZ selectively enters this cell type (Zhang et al., 2012) and only
results in β-cell depletion leaving other islet cell types intact. Additionally, other cells of the islet,
such as the α-cells proliferate after STZ treatment further demonstrating the specificity of STZ to
the β-cells (Li et al., 2000). STZ treated mice develop insulitis from β-cell death and survive
1 – 2 months after the onset of hyperglycemia. However, the type of cell death differs depending
on the dose of STZ administered. β-cell death can either occur by necrosis or apoptosis by two
mechanisms; (i) a direct toxic effect on the cell and; (ii) a delayed immune infiltration of the
islet. To cause β-cell death by apoptosis, multiple low dose STZ injections are administered.
Concentrations vary from 40 – 50 mg/kg and are administered daily for 5 days (Li et al., 2003;
O'Brien et al., 1996). Generally, apoptotic cell number is highest between the last day of STZ
injections and 48 hours afterwards (Li et al., 2003; O'Brien et al., 1996). Additionally, a second
peak of apoptosis 6 days after the last STZ injection is associated with β-cell death a peak of
immune infiltration (O'Brien et al., 1996). Initially, the low dose of STZ increases in DNA
fragmentation (Morgan et al., 1994) and nitric oxygen radicals (Szkudelski, 2001), eventually
causing DNA damage and cell death. Damaged and dying cells then present self-antigens, which
24
the immune system are not tolerant to causing immune mediated β-cell death (O'Brien et al.,
1996). Conversely, single high dose STZ injections, approximately between 100 – 200 mg/kg,
cause β-cell death by necrosis through direct cytotoxicity of STZ. These mice become
hyperglycemic within 48 hours of the last injection (Wu and Huan, 2008), whereas multiple low
dose STZ injections causes hyperglycemia within 1 – 5 days after the last injection (varies
depending on mouse age, gender and strain) (O'Brien et al., 1996)
The pathology in non-obese diabetic mice is initiated by immune infiltration of the islets
by T-cells, B-cells and lymphocytes causing insulitis, eventually leading to β-cell death (Leiter et
al., 1987; Sakata et al., 2012). These mice progressively develop diabetes, between 16 – 20
weeks for females and 21 – 28 weeks for males (Leiter et al., 1987). These mice are able to
survive without exogenous insulin 1 – 2 months after detection of hyperglycemia (Leiter et al.,
1987). Additionally, the progression of the disease in this model closely resembles the
pathogenesis of human type 1 diabetics because of similar immune infiltration, and is therefore a
popular model to investigate the mechanisms behind this disease. This model is more
physiologically relevant in terms of human disease, because β-cell death is spontaneous and
primarily caused by immune mediated cell death, rather than exposure to a toxin. However, STZ,
although not identical to the pathogenesis in human disease, is considered an appropriate model
to investigate the role of genes in the context of diabetes.
Peroxisome proliferator activated receptor gamma (PPARγ) co-activator 1-alpha (PGC-1α)
PGC-1α is a nuclear transcriptional co-activator that is responsible for oxidative
metabolism and mitochondrial biogenesis (Handschin and Spiegelman, 2006). PGC-1α is of
25
interest to study in diabetes because β-cell function relies greatly on functional mitochondria.
Mitochondria are necessary to metabolize glucose and promote insulin secretion (Maechler et al.,
2010). In fact, mitochondrial dysfunction is associated with the progression of the disease (Sivitz
and Yorek, 2010). PGC-1α was discovered in brown adipose tissue (BAT), by a yeast-two-
hybrid screen to identify PPARγ interacting proteins (Puigserver et al., 1998). PGC-1α is
robustly induced in BAT upon cold exposure, can be activated by β-adrenergic signalling in
brown fat, and regulates the expression of UCP-1 (uncoupling protein-1; a classical brown fat
marker and regulator of heat dissipation) (Puigserver et al., 1998). In addition to PGC-1α, two
other family members have been described; PGC-1β (Lin et al., 2002a) and PRC (PGC-related
co-activator) (Andersson and Scarpulla, 2001). PGC-1α has been associated with increased risk
of developing diabetes (Ek et al., 2001; Hara et al., 2002) and is dyregulated in the muscle and
islets of type 2 diabetics (Mootha et al., 2003; Olsson et al., 2011).
PGC-1α is approximately a 113 kilo Dalton (kDa) protein and contains an activation
domain, repression domain, arginine and serine (RS) rich domain and a RNA binding domain
(Figure 3) (Lin et al., 2005). PGC-1α’s co-activator properties arise from its ability to bind
histone acetlytransferase (HAT) containing proteins such as CBP (CREB binding protein), p300
and SRC-1 (steroid receptor coactivator-1) in the activation domain (Puigserver et al., 1999).
These interactions open heterochromatin, allowing for active transcription of genes.
Furthermore, PGC-1α can bind subunits of the mediator complex (in the RS and RNA binding
domain), which associates with RNA polymerase II, thus assisting in transcriptional initiation
(Wallberg et al., 2003). Moreover, through its LXXLL motifs, PGC-1α is able to bind a variety
of different nuclear receptors (Puigserver et al., 1998), such as PPARγ (Puigserver et al., 1998),
ERRα (estrogen-related receptor α) (Mootha et al., 2004) and NRF-1,2 (nuclear respiratory
26
factor-1,2) (Mootha et al., 2003; Wu et al., 1999). In addition, PGC-1α binds and activates
transcription factors that are not nuclear receptors, such as MEF2 (myocyte enhancer factor-2)
(Lin et al., 2002b) and FOXO1 (forkhead box protein O1) (Puigserver et al., 2003). Finally,
PGC-1α can be phosphorylated in its repression domain by Sirt1, a protein de-acetylase, to
repress its coactivator functions (Lin et al., 2005).
Co-activation of ERRα and NRF-1,2 by PGC-1α regulates mitochondrial biogenesis and
mitochondrial gene expression (Evans and Scarpulla, 1990; Mootha et al., 2004; Schreiber et al.,
2004; Virbasius et al., 1993). NRF-1 and -2 are able to bind a variety of mitochondrial genes,
including cytochrome c oxidase subunit 4 (COXIV), β-ATP synthase (ATP5B), CytoC and
mitochondrial transcription factor A (Tfam) (Evans and Scarpulla, 1990; Virbasius et al., 1993).
PGC-1α is a potent co-activator of NRF-1 and -2, and when ectopically expressed in muscle,
gene expression of these nuclear transcription factors increases and activates their downstream
targets (Wu et al., 1999). Therefore, PGC-1α is an important regulator of mitochondrial gene
expression. Additionally, PGC-1α’s ability to increase mitochondrial biogenesis is blocked by
NRF-1 inhibition (Wu et al., 1999). Furthermore, co-activation of ERRα by PGC-1α also
initiates mitochondrial biogenesis (Mootha et al., 2004; Schreiber et al., 2004) and this
interaction acts upstream of NRF-1 activation (Mootha et al., 2004).
Due to its potential role in regulating mitochondrial function, PGC-1α was first identified
as a master regulator of adaptive thermogenesis in BAT. Ectopic expression of PGC-1α in
murine BAT cells causes an increase of UCP-1, mitochondrial respiratory enzyme subunits,
COXII (cytochrome oxidase subunit II) and COXIV, and stimulates mitochondrial biogenesis as
indicated by an increase in mitochondrial DNA content (Puigserver et al., 1998). Additionally,
expression of PGC-1α in white adipose tissue (WAT) increases expression of UCP-1 and
27
mitochondrial biogenesis, creating a brown-fat-like phenotype of the WAT (Fisher et al., 2012;
Puigserver et al., 1998).
The ability of PGC-1α to bind a variety of transcription factors allows it to have different
physiological functions in a number of different tissues. PGC-1α is also induced upon fasting in
the liver. During fasting, it is critical to increase gluconeogenesis and shift fuel utilization from
glucose to fatty acids in order to maintain glucose homeostasis. PGC-1α expression regulates
these metabolic responses in the liver by activating hepatic transcription factors, such as FOXO1
(Puigserver et al., 2003) and HNF4α (hepatocyte nuclear factor 4 alpha) (Rhee et al., 2003) to
promote gluconeogenesis and PPARα to promote fatty acid oxidation (Vega et al., 2000).
PGC-1α interaction with FOXO1 increases the expression of Pck1 (phosphoenolpyruvate
carboxykinase 1) and G6pc (Glucose-6-phosphatase) (Puigserver et al., 2003); both critical
enzymes in the gluconeogenic pathway. Additionally, PGC-1α also acts via HNF4α to induce
expression of G6pc, promoting expression of another gluconeogenic gene, PEPCK
(Phosphoenolpyruvate carboxykinase) (Rhee et al., 2003). Furthermore, Vega et al. observe that
PGC-1α and PPARα cooperatively induce fatty acid oxidation genes, such as CPT1 (Carnitine
palmitoyltransferase I), MCAD (medium chain acyl-coenzyme A dehydrogenase) and LCAD
(long chain acyl-coenzyme A dehydrogenase) (Vega et al., 2000). Therefore, PGC-1α plays a
critical role in the liver to activate metabolic programs to maintain the amount of circulating
glucose and switch fuel consumption to fatty acids to ensure glucose homeostasis during a period
when energy intake is limited.
Furthermore, PGC-1α regulates metabolism in skeletal muscle. Here, PGC-1α is induced
during exercise and by β-adrenergic agonists (Baar et al., 2002). Expression in muscle can
28
promote the switch of muscle fibers from a fast-twitch phenotype, to a slow-twitch fiber
phenotype by binding to MEF2. These fibers are more oxidative and are associated with
endurance training (Lin et al., 2002b; Millay and Olson, 2013). PGC-1α serves as a sensor for
external cues to adapt skeletal muscle to exercise and its metabolic needs.
Additionally, PGC-1α plays an important role in the heart, brain and intestines. PGC-1α
is expressed abundantly in the heart, likely due to the heart’s great demand for ATP. Over-
expression of PGC-1α in cultured cardiomyocytes induces expression of mitochondrial genes
(MCAD and CPT-1) and stimulates mitochondrial biogenesis (Lehman et al., 2000). Conversely,
loss of PGC-1α in the heart leads to cardiac dysfunction (Arany et al., 2005; Leone et al., 2005).
The role of PGC-1α in the brain is less understood; however, a whole body knock out of PGC-1α
causes behavioural abnormalities including hyperactivity (Leone et al., 2005). These defects are
due to axonal degradation in the brain (Lin et al., 2004), which is thought to be caused by an
increase of ROS due to mitochondrial dysfunction (Lin et al., 2005). Finally, PGC-1α is an
important regulator in intestinal epithelium. Ectopic expression of PGC-1α induces expression of
key mitochondrial genes, such as Tfam, MCAD and ATP5B and an increase in mitochondrial
biogenesis as seen by an increase in mitochondria DNA copy number (D’Errico et al., 2001). An
increase in mitochondrial activity in the intestinal epithelium is necessary to maintain their
constant turnover, so PGC-1α is important in this process and also regulates cell fate (D’Errico et
al., 2011). Even though PGC-1α confers different functions in a variety of tissues, almost all of
these ascribed functions are related to an increase in metabolic activity of the tissue, an increase
in mitochondrial function and number, and are important for sensing changes (i.e. nutrient
availability) in the extracellular environment.
29
PGC-1 and -cell Biology
While one family member, PGC-1α, was shown to be expressed in β-cells, less is known
about the role of PGC-1α in β-cells. Interestingly, PGC-1α is increased in β-cells of rodents in
the context of diabetes (Yoon et al., 2003). Previous studies investigating PGC-1α function in
β-cells have used anti-sense oligonucleotides (ASOs) or short-interfering RNA (siRNA) to
knock-down co-activator expression or adenovirus for over-expression (De Souza et al., 2005;
De Souza et al., 2003; Kim et al., 2009; Yoon et al., 2003). There is only one genetic study,
which utilizes a tetracycline-dependent inducible gene system in vivo expression to enable
PGC-1α over-expression specifically in β-cells (Valtat et al., 2013).
Targeting PGC-1α by ASOs achieved an 80% knock down, which decreased blood
glucose in a diet-induced diabetes model in mice. Additionally, these mice exhibited an increase
in serum insulin levels, were more glucose tolerant and more insulin sensitive (De Souza et al.,
2005). The same group demonstrated that PGC-1α inhibition (using ASOs) reversed cold-
inhibition of insulin secretion and prevented UCP-2 expression in islets upon cold exposure
(De Souza et al., 2003). However, it is important to note that ASOs in vivo also target other
organs (i.e. liver and adipose tissue) very efficiently and it is impossible to dissociate the
contribution of PGC-1α knock-down in these other organs to the phenotype.
The role of PGC-1α in glucolipotoxicity-associated β-cell dysfunction was evaluated by
siRNA. Glucolipotoxic (high glucose, high lipid) environments decrease β-cell insulin content
and increase endogenous expression of PGC-1α (Kim et al., 2009; Zhang et al., 2005). Insulin
content was decreased by both glucolipotoxicity and over-expression of PGC-1α using
adenovirus (Ad-PGC-1α) in cultured rat islets. This effect on insulin content was reversed by
30
adenoviral expression of siPGC-1α (Ad-siPGC-1α) (Kim et al., 2009). Since decreasing PGC-1α
was beneficial for β-cell health in vitro, they also investigated whether reduced PGC-1α
expression protects against β-cell dysfunction in vivo. Using 90% pancreatectomized mice to
induce hyperglycemia, they delivered Ad-siPGC-1α to the islets by the celiac artery. Mice
injected with Ad-siPGC-1α are moderately more glucose tolerant, have lower fasting glucose and
higher fasting insulin levels (Kim et al., 2009). Consistent with this study, over-expressing
PGC-1α by adenovirus in cultured mouse islets in vitro negatively effects glucose stimulated
insulin secretion and these islets cannot rescue STZ induced hyperglycemia through
transplantation (Yoon et al., 2003). These studies demonstrate that reduced PGC-1α expression
in rodent islets positively impacts β-cell function, whereas PGC-1α over-expression is
detrimental for functionality. However, adenoviral delivery via the celiac artery is not β-cell
specific and therefore the effects of PGC-1α knock-down or over-expression can be confounded
by off target effects. These caveats demonstrate the need to manipulate PGC-1α expression in
genetic models specifically in β-cells.
Mice expressing PGC-1α downstream of a tetracycline-dependent promoter (TetO)
crossed with mice expressing the tetracycline transactivator (tTA) under the Insulin1 gene
promoter allows β-cell specific over-expression of PGC-1α that can be turned off in the presence
of doxycycline (DOX). Over-expression of PGC-1α in β-cells during mouse development and
throughout life causes hyperglycemia, lower fed insulin serum levels, glucose intolerance and
altered glucose-stimulated insulin secretion (GSIS) in 6-month old mice. These mice also have
reduced β-cell mass and reduction of insulin content. In mice with increased β-cell PGC-1α only
during adulthood (given DOX from birth until 4 months of age to prevent PGC-1α
over-expression) have normal glucose tolerance and GSIS at the same age, suggesting that the
31
effects of PGC-1α on β-cell function are caused by neonatal over-expression of the gene
(Valtat et al., 2013). This demonstrates that PGC-1α over-expression in vivo specifically in
β-cells negatively impacts β-cell function only when over-expressed neonatally.
The literature for PGC-1α function in β-cells is limited; however, these rodent studies
complement each other. Simply these studies have shown that a knock-down of PGC-1α in
β-cells increases β-cell function and over-expression impairs it. This data suggests that
PGC-1α expression could be a contributing factor to diabetes and inhibiting it could be a possible
therapy by improving β-cell function. However, PGC-1α expression is decreased in human type
2 diabetic patients (Ling et al., 2008), raising the possibility that loss of PGC-1α a may have yet
unidentified pathological effects on β-cell health. The recent discovery of novel PGC-1α
isoforms adds and additional layer of complexity making it of interest to elucidate any potential
roles of these isoforms in β-cell physiology.
PGC-1α Isoforms
In addition to the canonical PGC-1α transcript (herein referred to as PGC-1α1), there
recently has been the emergence of new isoforms. The first isoforms published were PGC-1α-b
and PGC-1α-c (Figure 4) (Miura et al., 2008). These differed from canonical PGC-1α only in
their N-terminus, with a 16 amino acid difference due to alternative splicing of exon 1 (Miura et
al., 2008). This was the first report of a novel exon1 of PGC-1α, found approximately 14
kilobases (kb) upstream of the canonical exon 1. This novel exon is alternatively spliced to a
shared exon 2 giving rise to two new isoforms (Miura et al., 2008).
32
Recombinant PGC-1α-b and PGC-1α-c are functional as transcriptional co-activators in
vitro and in vivo. Fusion of the isoform cDNA to Gal4-DB (DNA binding domain) constructs
and transfection into HEK293 cells activated a UAS-luciferase reporter (Miura et al., 2008).
Furthermore, transgenic mice over-expressing PGC-1α-b and PGC-1α-c in muscle, show
increased expression of classical PGC-1α regulated genes, such as Tfam, COXII, COXIV and
MCAD. Additionally, after injection of mouse muscle with clenbuterol (a β-adrenergic activator)
or exercise, total PGC-1α expression increased, which induces, PGC-1α1, PGC-1α-b and
PGC-1α-c (Miura et al., 2008). It was later shown that PGC-1α-b is induced upon exercise in
human muscle, and constitutes approximately 10% of the total PGC-1α transcripts after 2 hours
of exercise (Norrbom et al., 2011).
PGC-1α-b and PGC-1α-c were also identified in liver and brown adipose tissue (BAT)
using primers that detect only the exon 1 sequence. Unlike PGC-1α1, these isoforms are not
induced in the liver upon fasting; however, they are moderately increased upon cold exposure in
the BAT, like PGC-1α1.
Similar to PGC-1α-b and PGC-1α-c, two other isoforms were recently published,
PGC-1α2 and PGC-1α3 (Figure 4) (Chinsomboon et al., 2009; Ruas et al., 2012). PGC-1α2 and
PGC-1α3 have the same upstream exon 1 as PGC-1α-b- and PGC-1α-c, respectively (Figure 4);
however, this group identified a novel alternative promoter. This promoter lies approximately
14kb upstream of the proximal promoter, similar to the distance where the alternative exon 1 is
found for PGC-1α-b and PGC-1α-c (Miura et al., 2008). While PGC-1α-b and PGC-1α-c were
not originally described as being transcribed from this new alternative promoter, with its
discovery it is likely that these isoforms are regulated by this site. However, PGC-1α-b / PGC-
1α-c and PGC-1α2 / PGC-1α3 differ in the majority of their exons downstream of the novel
33
exon 1 (Figure 4). Chinsomboon et al. noted that PGC-1α2 and PGC-1α3, similar to PGC-1α-b
and PGC-1α-c, were expressed relatively abundantly in muscle and BAT, but absent from other
metabolically active tissues, such as the liver (Chinsomboon et al., 2009). In contrast to
PGC-1α-b and PGC-1α-c, they also demonstrated that after exercise and β-adrenergic activation,
the total amount of PGC-1α expressed is attributed only to isoforms from the alternative
promoter. Since Miura et al. used primers specific for exon 1 to characterize PGC-1α-b and
PGC-1α-c, which also recognize PGC-1α2 and PGC-1α3 respectively, it is possible that
expression changes reported could be in part due to changes in PGC-1α2 and PGC-1α3.
Regardless of the differential expression, the functions of PGC-1α2 and PGC-1α3 are unknown
and those of PGC-1α-b and PGC-1α-c are closely related to that of canonical PGC-1α1.
Another recently identified isoform, NT-PGC-1α, is transcribed from the proximal (or
canonical) promoter and encodes a truncated version of full-length original PGC-1α (Zhang et
al., 2009). This isoform was discovered while cloning full length PGC-1α from mouse BAT
cDNA library. It was found to have a 31 basepair (bp) inclusion in intron 6 causing an in frame
premature stop codon in the transcript, resulting in translation of a truncated version of PGC-1α
(Figure 4). NT-PGC-1α is induced in BAT upon cold exposure and the liver upon fasting.
NT-PGC-1α induces expression of UCP-1 and CPT1 in cultured adipocytes and mitochondrial
biogenesis in cultured brown adipocytes (Zhang et al., 2009). Alternative splicing of intron 6
does not favour one transcript over the other in white adipose tissue (WAT), spleen or heart;
however, splicing appears to favour NT-PGC-1α expression in the brain due to an unknown
mechanism (Zhang et al., 2009). Authors also note that, although mRNA levels are similar in
tissues examined, protein expression of NT-PGC-1α is more robust compared to PGC-1α1. This
is attributed to decreased susceptibility for proteosomal degradation. Although NT-PGC-1α
34
seems physiologically regulated similarly to PGC-1α1 at the level of transcription, it has unique
properties such as increased protein stability (Zhang et al., 2009).
Up to this point, all biologically active PGC-1α isoforms regulate the same transcriptional
pathways as PGC-1α1 and have closely related functions surrounding mitochondrial metabolism.
However, in 2012, Ruas et al. published the first report of a novel isoform, PGC-1α4, having a
distinct function to PGC-1α1 in muscle (Ruas et al., 2012). Like NT-PGC-1α, PGC-1α4 is
truncated due to the introduction of an in frame stop codon in intro 6, however, PGC-1α4 is
transcribed from the alternative promoter (Figure 4) and thus differs in its exon 1 sequence
(Figure 5).
A microarray screen of myotubes adenovirally over-expressing PGC-1α1, PGC-1α2,
PGC-1α3, and PGC-1α4, revealed that PGC-1α1 and PGC-1α4 regulate only 98 genes similarly,
with the majority of genes regulated by either isoform being unique. From this screen, they
discovered that PGC-1α4 does not regulate classical PGC-1α genes, such as CytoC, COXVb
(cytochrome c oxidase subunit V b), Glut4 (glucose transporter type 4), but does regulate genes
involved in cell growth and proliferation of muscle, such as IGF1 (insulin-like growth factor 1)
and MSN (myostatin). Furthermore, in vitro adenoviral over-expression of PGC-1α4 causes
muscle fiber hypertrophy and knock-down of PGC-1α4 prevents this action in response to
clenbuterol. Additionally, mice over-expressing PGC-1α4 in muscle either by adenovirus,
plasmid injection or in a transgenic model, have increased muscle size and strength or decreased
muscle wasting in limb suspension and cancer cachexia models (Ruas et al., 2012).
In muscle, PGC-1α1 is responsible for mitochondrial biogenesis and oxidative
metabolism in response to exercise, while PGC-1α4 is responsible for muscle hypertrophy and
35
strength (Millay and Olson, 2013). PGC-1α4 transcripts have been identified in multiple
metabolically active tissues and is the only PGC-1α isoform thus far shown to have a unique
function.
PGC-1α and Disease
PGC-1α dysregulation is linked to diseases particularly in regards to mitochondrial
dysfunction. Neurogenerative diseases such as Parkinson’s, Alzheimer’s and Huntington’s, are
associated with a decrease in mitochondrial function and oxidative phosphorylation (OXPHOS)
genes (Schon and Manfredi, 2003). Low PGC-1α is also linked to heart disease (Arany et al.,
2005) and hepatic porphyrias in humans (Handschin et al., 2005). Furthermore, loss of PGC-1α
expression in mice causes hepatic steatosis in the liver (Estall et al., 2009; Leone et al., 2005)
and PGC-1α expression is decreased in the livers from patients with varying degrees of non-
alcoholic fatty liver disease (NAFLD) (Westerbacka et al., 2007). Moreover, a common PGC-1α
glycine to serine substitution polymorphism is correlated to NAFLD in children (Lin et al.,
2013).
PGC-1α dysregulation is also strongly correlated to type 2 diabetes; however, it is
currently debated whether loss of PGC-1α is a cause or consequence of the disease. PGC-1α
dysregulation in muscle has been linked to being both a cause (Schuler et al., 2006) and
consequence (Shiff et al., 2009) of diabetes, demonstrating the lack of consensus in the field. In
human type 2 diabetics, OXPHOS gene expression is decreased in muscle from these patients,
which correlates with a 20% decrease in PGC-1α transcripts (Mootha et al., 2003). Decreased
OXPHOS gene expression also correlates with mitochondrial dysfunction in diabetic patients
36
and is speculated as a potential cause of insulin resistance in the muscle (Mootha et al., 2003).
OXPHOS genes are also decreased in islets of type 2 diabetics (Olsson et al., 2011) and PGC-1α
expression is low in comparison to control healthy islets (Ling et al., 2008). Moreover
knock-down of PGC-1α in human islets prevents insulin secretion (Ling et al., 2008).
Interestingly, these findings in human β-cells contradict both the knock-down and over-
expression studies in mouse models of PGC-1α. This could be attributed to the fact that PGC-1α
is up-regulated in β-cells of obese rodents (Yoon et al., 2003), whereas in diabetic humans it is
down-regulated (Ling et al., 2008). This illustrates the differences between rodents models of
diabetes and humans and why it is important to eventually study genes of interest in human
tissues.
Interestingly, a polymorphism of PGC-1α is also associated with an increased risk in
developing type 2 diabetes. A Gly482Ser amino acid substitution is found in various populations,
including Danish (Ek et al., 2001) and Japanese populations (Hara et al., 2002); however, this
mutation is ethnicity specific, since there was no correlation between this polymorphism and
incidence of diabetes in French or Austrian populations (Lacquemant et al., 2002; Oberkofler et
al., 2004). Even though the Gly482Ser PGC-1α variant is associated with type 2 diabetes (Ek et
al., 2001), it has not been associated with body mass index (BMI), fasting glucose or fasting
insulin (Barroso et al., 2006). Currently it is unclear whether PGC-1α dysfunction in diabetes is
restricted to the canonical PGC-1α1 transcript or also involves the function of other PGC-1α
isoforms. However, it is clear that the dysregulation of PGC-1α can contribute to the
pathogenesis of diabetes.
While most assays to characterize reduced PGC-1α expression (typically qPCR) in
human disease collectively measure all isoforms of PGC-1α, until recently, characterization of
37
the role of PGC-1α in disease only focused on the canonical full-length transcript. In mice,
PGC-1α over-expression in muscle protects against declining mitochondrial function and
increased oxidative damage during aging and improves insulin sensitivity (Wenz et al., 2009).
Over-expression of canonical PGC-1α1 inhibited insulin secretion in rodent β-cells (Valtat et al.,
2012), while reduced levels seen in diabetic islets from humans include all isoforms (Ling et al.,
2008). Furthermore, whole body PGC-1α transgenic mice have hepatic insulin resistance, but
improved muscle insulin sensitivity, lower ROS and NF-κB signalling in muscle (Liang et al.,
2009). In terms of PGC-1α isoforms, only PGC-1α4 has been investigated within the context of
disease and can protect against cancer-induced cachexia in muscle (Ruas et al., 2012).
Objectives
β-cell dysfunction and death are hallmarks of both type 1 and type 2 diabetes. PGC-1α
dysregulation is associated with diabetes. OXPHOS genes controlled by PGC-1α are down in
muscle (Mootha et al., 2003) and islets (Olsson et al., 2011) and a polymorphism of PGC-1α is
associated with an increased risk of developing T2D (Ek et al., 2001; Hara et al., 2002). PGC-1α
transcripts are decreased in islets of diabetic humans (Ling et al., 2008); however, it is still
unknown whether PGC-1α isoforms play a role in the pathogenesis of diabetes.
We set out to investigate whether PGC-1α isoforms transcribed from the alternative
promoter played a role in the pathogenesis of diabetes. Our main objectives were; (i) to
determine whether PGC-1α isoforms are differentially regulated in β-cells; (ii) investigate
whether these isoforms have unique molecular functions and; (iii) determine whether their
dysregulation contributes to the pathogenesis of diabetes.
38
SECTION II:
MATERIALS AND METHODS
39
Cell Culture
INS-1 parental cells, between passage 8 to 25, were maintained in RPMI 1640 (Wisent)
with 10% FBS (Wisent), 1% penicillin and streptomycin (Wisent), and 1x supplement (10mM
HEPES, 1mM sodium pyruvate, 50μM β-mercaptoethanol). INS-1 cells were seeded into 6-well
plates at a density of 0.45x106 cells/ml the evening prior to experimentation, unless otherwise
stated. Primary islets isolated from wild-type male and female C57Bl/6J mice of various ages,
were cultured in RPMI 1640 with 10% FBS and 1% penicillin and streptomycin. Islets were
cultured for 2 days prior to experimentation to ensure the death of extraneous acinar tissue and
immune cells. All were maintained at 37 C with 5% CO2.
Islet Isolation
Pancreata from male and female C57Bl6 mice, between 20 – 30 g, were perfused with
0.4 U/mL of Liberase TL enzyme (Roche). Pancreata were digested at 37˚C for 30 minutes. They
were disrupted by vigorous shaking by hand 45 times, then mixed with HBSS solution (Wisent)
with 0.1% bovine serum albumin and 0.02 M HEPES and centrifuged for 3 minutes at 500 g at
4˚C. After 4 washes in HBSS solution, samples were resuspended in 9 mL Histopaque
(density 1.077gml, Sigma), and layered with RPMI 1640 without glucose (Wisent). To separate
the islets based on this density gradient, samples were centrifuged for 30 minutes at 400 g at 4˚C.
Supernatant was decanted and total volume was completed with HBSS without BSA and
HEPES. Samples were centrifuged for a final time for 3 minutes at 500 g at 4˚C, resuspended in
10 mL of RPMI 1640 with 10% FBS and 1% penicillin and streptomycin and picked into
untreated plates.
40
RNA Extraction and cDNA Synthesis
Total RNA from INS-1 cells was extracted using Trizol reagent (Invitrogen), as indicated
by the manufacture’s protocol. Total RNA from primary isolated islets (minimum 100 islets) was
extracted using the RNeasy Micro Kit (Qiagen), as indicated by the manufacturer’s protocol.
For cDNA synthesis, first, 1 µg of RNA from INS-1 cells and 400 ng of RNA from
isolated islets, were incubated with 1 U/mL DNase at 37˚C for 15 minutes, followed by 15
minutes at 65˚C for DNase heat inactivation. Then, total RNA in a total volume of
20 µl was reverse transcribed with 50 U MultiScribe™ reverse transcriptase (Life Technologies)
and 20 U RNase Inhibitor (BioBasic). cDNA was synthesized at 25˚C for 10 minutes, 37˚C for
120 minutes and 85˚C for 5 minutes. 80 µl of water (1:5 dilution) was added to each sample and
was stored at -20˚C. cDNA samples were assessed by qPCR.
Quantitative Real-Time PCR
cDNA was subjected to two amplifications, one for the gene of interest and one for the
endogenous control hypoxanthine-guanine phosphoribosyltransferase (HPRT). Each condition
was in triplicate, in 5 µl reactions in a 384-well plate using Power SYBR®
Green PCR master
mix (Life Technologies). The cycling program was in two steps, a polymerase activation step for
2 minutes at 50˚C and 10 minutes at 95˚C, followed by 40 cycles of 15 seconds at 95˚C and 1
minute of 60˚C, using the Viia7 system from Life Technologies. Data was normalized to the
endogenous control and relative mRNA expression was determined using the ΔΔCt method.
Graphpad Prism was used for graphing results and performing statistical analysis.
41
Protein Analysis
Cells were lysed in RIPA buffer (50mM Tris, 150mM NaCl, 1% NP-40, 0.5% sodium
deoxycholate, 0.1% SDS) plus protease cocktail inhibitor (Calbiochem) and protein
concentrations were estimated by DC Assay. Equal amounts of total protein (40-60 µg) were
resolved by SDS-PAGE on 10% poly-acrylamide gels for PGC-1α immunoblots and 12% gels
for cleaved caspase-3 immunoblots, and were transferred onto polyvinylidene difluoride (PVDF)
membranes (GE Healthcare). PGC-1α isoforms were detected using the anti-PGC-1α mouse
4C1.3 antibody (1:1000, Millipore™) and activation of caspase-3 was detected using a cleaved
caspase-3 antibody (1:500, Cell Signalling). A β-actin (1:50000) antibody was used as a loading
control. Both, PGC-1α and cleaved caspase-3 primary antibodies were diluted in 2% milk and
membranes were incubated at 4˚C overnight. The primary β-actin antibody was diluted in Tris
buffered saline with 0.1% Tween (TBST) and incubated for 1 hour at room temperature.
Secondary antibodies used were goat anti-mouse IgG antibody conjugated to horseradish
peroxidise (HRP) (1:10000, GenScript) and goat anti-rabbit IgG antibody conjugated to HRP
(1:5000, GenScript), for the PGC-1α and cleaved caspase-3 antibodies, respectively. Signal
detection was performed using an ECL detection system (GE Healthcare©
). Films were exposed
for 1 hour, except for β-actin, which was exposed for 1 minute or less.
Genotyping
Genotyping of mice with PGC-1α floxed alleles with or without the cre transgene was
performed on DNA prepared from tails of 3 week old mice. DNA was extracted by boiling the
tails at 100˚C in 200 µl of 50 nM sodium hydroxide (NaOH) for 20 minutes. After 20 minutes,
42
20 µl of 1M TRIS pH 6.8 was added to the tail digested and vortexed for 15 seconds. They were
frozen and stored at -20˚C.
PGC-1α floxed alleles were genotyped using forward (5’- TCCAGTAGGCAGAGATTT
ATGAC -3’) and reverse (5’- TGTCTGGTTTGACAATCTGCTAGGTC -3’) primers which
flank the 5’ loxP site. Genotyping was performed by polymerase chain reaction (PCR) for 35
cycles at 96˚C for 1 minute, 58˚C for 1 minute and 72˚C for 1 minute. Expected band patterns
are; one band of 400 bps for the mutant; 1 band each of 400 bps and 360 bps for a heterozygote
and; 1 band of 360 bps for WT (Figure 6A). Cre recombinase positive alleles were genotyped
using a forward (5’- TAAGGGCCCAGCTATCAATGGGAA -3’) primer in the mouse insulin
promoter and a reverse (5’- GTGAAACAGCATTGCTGTCACTT -3’) primer in the Cre
transgene. Genotyping was performed by PCR for 35 cycles at 94˚C for 30 seconds, 60˚C for 1
minute and 72˚C for 1 minute. The expected band pattern is 1 band at 800 bps for Cre positive
mice (Figure 6B). Both genotyping reactions were visualized on 2% agarose gels stained with
Ethidium Bromide.
Induction of PGC-1α Isoforms
INS-1 and primary isolated islets were treated with forskolin (FK) (10μM) and Exendin-4
(Ex-4) (50nM) for 2 hours (for RNA isolation) or 4 hours (for protein isolation, to allow
sufficient time for mRNA transcription and protein synthesis), or DMSO or water as the vehicle,
respectively. Additionally, INS-1 cells were treated with cytokines (TNFα (50ng/ml), IFNγ
(50ng/ml) and IL-1β (10ng/ml)) or water as the vehicle, for 12 hours prior to harvesting cells for
RNA or protein isolation. qPCR was performed using isoform specific primers to identify the
43
relative mRNA expression of each and protein analysis using an anti-PGC-1α antibody
(Millipore).
Cleaved Caspase-3 Assay
INS-1 cells were infected with adenoviruses expressing GFP, PGC-1α1 and PGC-1α4,
previously described in (Ruas et al., 2012), with a titer of 0.625 x 107 ifu/ml for 8 hours, after
which the virus media was changed for fresh media. 30 hours post infection, cells were treated
with cytokines for 18 hours, for a total of 48 hours. Total protein was isolated in RIPA buffer and
the same protein lysates were resolved in 2 separate SDS-PAGE gels, one to visualize PGC-1α
isoform expression and the other for cleaved caspase-3.
PGC-1α Knock Out and Streptozotocin in vivo Experiment
PGC-1α floxed/floxed MIP-Cre (PGC-1αfl/flCre+
) and PGC-1α floxed/floxed (PGC-1αfl/fl
)
(age/gender-matched littermate controls) mice, on a mixed background, were gavaged with
tamoxifen (100mg/kg, Sigma Aldrich) for 10 days, separated by a 2 day rest in between. After 2
weeks recovery to allow complete gene excision and elimination of tamoxifen, mice were
injected intraperitoneally (ip) with Streptozotocin (STZ) (Bioshop) in 0.1mM sodium citrate
pH 4.5, at a dose of 50mg/kg everyday for 5 days. Random fed blood glucose measurements
were taken bi-weekly after STZ injections to monitor blood glucose over time. 1 day post-final
STZ injection, select mice were sacrificed (n=7-8) and the pancreata were harvested and fixed in
4% PFA for histology, for PGC-1αfl/fl
, PGC-1αfl/flCre+
, WT and MIP-Cre only mice, to control for
possible effects of the cre-recombinase transgene. These samples were embedded in paraffin and
sectioned (5 μm) for immunohistochemical staining for cleaved-caspase-3. 11 days post-STZ
44
injection, an early OGTT (n=9 PGC-1αfl/fl
, n=8 PGC-1αfl/flCre+
) was performed following an
overnight fast and gavaged with 1g/kg of glucose. 18 days post-final STZ injections, glucose
measurements and blood samples were taken via tail vein following an overnight fast and then
after re-feeding for 2 hours. Lastly, a late OGTT (n=6 PGC-1αf/lfl
, n=5 PGC-1αfl/flCre+
) was
performed, 32 days after the last STZ injection following an overnight fast and were gavaged
with 1g/kg of glucose. All glucose measurements were taken using the FreeStyle Lite glucometer
(Abbot Diabetes Care). Mice were sacrificed one week after and pancreata were fixed in 4%
paraformaldehyde. Mice were maintained on a chow diet throughout the duration of the
experiments. Mice were housed in a mouse-specific pathogen-free (SPF) facility, with a 12 hour
light and dark cycle. All mouse experiments were approved by the Institut de recherches
cliniques de Montréal Animal Care Committee (protocol number 2011.14) and complied with
the Canadian Council of Animal Care rules.
Insulin Enzyme-Linked Immunosorbent Assay (ELISA)
Serum insulin concentration was determined using an ultrasensitive ELISA (Alpco),
according to the manufacturer’s protocol. Serum was collected from the mice via the tail vein,
into 0.4 U/mL of heparin. Serum was separated from blood cell fraction by centrifugation at
15,000 rpm for 5 minutes. 25 μl of serum was used per well for the insulin ELISA. Standards
and controls recommended for 25 μl reactions were also used. Samples were incubated with
100 μl of the detection antibody at room temperature for 2 hours, shaking at 800 rpm. Samples
were read at 450 and 650 nm, to determine OD of the samples and of the background,
respectively. Data was analyzed using GraphPad Prism.
45
Immunohistochemistry
Immunohistochemistry (IHC) was performed on 5 µm thick paraffin sections. Before
staining, sections were de-waxed in Xylene 4 times for 5 minutes, followed by hydration in
decreasing concentrations of alcohol, from 100 to 70 percent, for 3 minutes. Sections were
quenched in endogenous peroxidase and antigen retrieval was performed using Rodent
Decloaker (Biocare) for 40 minutes at 85˚C, and blocked with Rodent Block M (Biocare) for 20
minutes. Sections were incubated with the cleaved caspase-3 antibody (Cell Signalling) diluted
1:150 in diluent solution, and were incubated overnight at 4˚C. The next day, the sections were
incubated with rabbit HRP polymer secondary antibody (Biocare) for 20 minutes followed by
colour developing using the DAB Chromogen kit (Biocare) for 3 minutes. Sections were
counterstained using Mayer’s hematoxylin, dehydrated using increasing concentrations of
alcohol from 70 to 100 percent and were rinsed in Xylene and mounted in mounting media.
Cleaved caspase-3 Quantification
One section for each mouse was stained for cleaved caspase-3 following the
immunohistochemistry protocol described above. Photos were taken using a Leica Axiophot
MZ12 with a 10x objective magnification. For each islet in the section the number of positively
and negatively stained cells were counted. Percentage of positive cells was calculated by
dividing the number of positive cells, by total number of cells in the islet multiplied by 100.
46
Statistical Analysis
Statistical analysis was calculated using GraphPad Prism. Results were expressed as
means ± SD for cell experiments and ± SEM for animal experiments and two-tailed student’s
t-test was used to determine p values, unless otherwise stated. Statistical significance was
defined at p < 0.05.
47
SECTION III:
RESULTS
48
PGC-1α isoforms are differentially expressed in a rodent β-cell line and primary islets
PGC-1α is an important modulator of mitochondrial function in metabolically active
tissues. Recently, Ruas et al., described the function of PGC-1α isoforms found in muscle and
reported that they are also expressed in other metabolically active tissues; however, it is
unknown whether these isoforms have other tissue-specific functions (Ruas et al., 2012).
Interestingly, total PGC-1α levels are decreased in islets of type 2 diabetic patients, but the
contribution of different PGC-1α isoforms to diabetes has not been reported, making it of interest
to elucidate whether differential expression or function of these isoforms occurs in β-cells and
contributes to the pathogenesis of the disease.
Since the expression of these novel isoforms has not been investigated in β-cells, the first
step in determining their importance in β-cell biology was to measure basal expression levels of
each of the isoforms identified in muscle, in an immortalized β-cell line (INS-1) and primary
islets isolated from mice. In particular, we wanted to investigate isoforms regulated by the
alternative promoter because one isoform from this promoter, PGC-1α4, has a unique function
from PGC-1α1. Since PGC-1α2 and PGC-1α3 are also transcribed form this promoter these
isoforms potentially also have novel functions in β-cells. Because of sequence similarity between
different isoforms, the primers used to detect PGC-1α2 and PGC-1α3 also detect PGC-1α-b and
PGC-1α-c, respectively, and PGC-1α4 primers can also detect NT-PGC-1α. So, all the isoforms
are measured at first to narrow down the isoforms of interest. For simplicity sake, however, only
PGC-1α1, PGC-1α2, PGC-1α3 and PGC-1α4 will be referred to, but the primers or antibody
cannot distinguish them at this time.
To determine whether the isoforms, PGC-α1 (the canonical transcript), PGC-α2, PGC-α3
and PGC-α4, were endogenously expressed in β-cells and can be regulated under physiological
49
conditions, INS-1 cells and primary islets were treated with 10 μM Forskolin, an activator of
adenylyl cyclase (linked to GPCR signalling) known to induce expression of the canonical
PGC-1α (Cowell et al., 2008), for 2 hours or with DMSO as the vehicle control. Under these
conditions, PGC-1α1, PGC-1α3 and PGC-1α4 were detected at the mRNA level in INS-1 cells
and primary islets (Figure 7A and 7B); however, PGC-1α2 expression was undetected by qPCR,
in either model, and the induction of PGC-1α3 by forskolin in islets did not reach statistical
significance (Figure 7B).
While PGC-1α3 mRNA was detected by qPCR only PGC-1α1 and PGC-1α4 proteins
were visualized by Western blotting after a 4 hour treatment with Forskolin, in both INS-1 cells
(Figure 7C) and primary isolated islets (Figure 7D). There was approximately a 5-fold and
12-fold induction of PGC-1α1 and PGC-1α4 in INS-1 cells, respectively (Figure 7E), but only a
~3-fold induction of these isoforms in primary islets (Figure 7F). For PGC-1α4, there was a more
dramatic induction in INS-1 cells after forskolin treatment (Figure 7E), but only a moderate
induction in islet cells (Figure 7F). The exact fold increase of PGC-1α4 in islets is difficult to
determine due to the high variability. Since islets are a more physiological model of β-cell
biology, it is likely that PGC-1α4 expression is more closely related to levels observed in islets
rather than INS-1 cells. However, since PGC-1α4 is expressed and regulated similarly in both
islets and INS-1 cells, they are a useful and relevant culture model to study isoform function.
These results show that, of the isoforms investigated, PGC-1α1 and PGC-1α4 are
endogenously expressed in β-cells, and can be regulated by cAMP in an isoform dependent
manner.
50
Cytokines modulate the expression of PGC-1α1 and PGC-1α4
The biological function of canonical PGC-1α1 has been extensively studied in multiple
cell types where it was found to be a master regulator of mitochondrial function and biogenesis.
However, only one function for PGC-1α4 has been described in muscle, that being induction of
muscle cell growth, a function not shared by the related isoform PGC-1α1.
We next set forth to determine whether PGC-1α1 and PGC-1α4 have unique functions in
β-cells. Unpublished microarray data from our lab using hepatocytes over-expressing either
PGC-1α1 or PGC-1α4 demonstrated that in the presence of an inflammatory stimulus (TNF-α)
PGC-1α4 differentially regulated genes involved in inflammatory pathways, compared to
PGC-1α1 and the control vector. This preliminary data in liver cells suggested that inflammatory
stimuli could affect the expression of the PGC-1α isoforms and the genes that they regulate.
Inflammatory signalling is known to have dramatic effects on the biology of β-cells and plays an
important role in the pathogenesis of diabetes. Since our preliminary screen suggested that the
differential gene regulation between PGC-1α1 and PGC-1α4 could be revealed upon exposure to
inflammatory stimuli, we next determined whether cytokines impacted isoform expression in
β-cells.
To determine whether activation of inflammatory signalling also regulates PGC-1α
isoform expression in β-cells, INS-1 cells were treated with a cytokine cocktail of TNF-α, IL-1β
and IFNγ, which have been shown to cause β-cell dysfunction with chronic exposure (Donath et
al., 2003). Indeed, the expression of PGC-1α1 and PGC-1α4, but not PGC-1α3, were induced at
the mRNA level in a time dependent manner (Figure 8A). MCP-1 is an early target of NF-κB
activity and its mRNA was measured as a positive control for the activation of inflammatory
signalling pathways (Figure 8A). PGC-1α2 was also measured; however, it was again
51
undetectable by qPCR. Expression of these two isoforms was most robust at 12 hours (~8 fold
induction for both PGC-1α1 and PGC-1α4) and then began to decline, to only a 5-fold and 6-fold
induction after 24 hours of cytokine stimulation, for PGC-1α1 and PGC-1α4, respectively
(Figure 8A). Interestingly, the induction of the PGC-1α isoforms was most robust after early
markers of inflammatory signalling began to decline, as determined by MCP-1 gene expression.
This suggests that the induction of PGC-1α1 and PGC-1α4 may not result from initial and direct
activation of inflammatory signalling pathways (i.e. NK-κB activation) in INS-1 cells, but may
increase as a compensatory response to the inflammatory reaction. This induction is unique for
PGC-1α1 and PGC-1α4, since PGC-1α2 expression was undetectable by qPCR and PGC-1α3
expression trended downward at all time points observed as compared to the control; however,
this was not statistically significant (Figure 8A). Additionally, an increase in PGC-1α4 protein
expression by 20-fold (Figure 8B, 8C), was observed. While PGC-1α1 mRNA was increased, the
protein was not detected by Western blotting. However, we could not rule out a meaningful
induction of PGC-1α1 at the protein level (Figure 8B) because the antibody is quite poor and
PGC-1α1 protein is known to be relatively unstable, with a half-life of 2-3 hours (Fernandez-
Marcos and Auwerx, 2011; Sano et al., 2007). Cytokine induction of the isoforms in islets
proved to be technically challenging. Cytokine responsiveness of the islets was different
compared to INS-1 cells, so concentrations and time-points still need to be optimized.
These results reveal that the expression of both PGC-1α1 and PGC-1α4 can be induced in
a time dependent manner in cultured β-cells in response to an inflammatory stimulus, but it is
likely due to an indirect, rather than direct mechanism. These findings indicate that PGC-1α
isoforms may play differential roles in the response to inflammatory cytokines in β-cells.
52
PGC-1α4 prevents the cleavage of caspase-3 in response to cytokines
The next step was to determine the physiological implications of cytokine regulation of
the PGC-1α isoforms. It is known that chronic exposure of β-cells to cytokines, particularly
TNFα, IL-1β and IFNγ, leads to β-cell dysfunction and death by apoptosis (Donath et al., 2003).
Cytokine signalling is also thought to mediate β-cell death in both type 1 and type 2 diabetics.
Since the basal expression level of PGC-1α3 was very low and the protein was not visible under
basal or Forskolin treated conditions we focussed on determining whether PGC-1α1 and
PGC-1α4 have an impact on in β-cell health and survival in response to cytokines.
Although many apoptotic pathways exist, the activation of caspase-3, also known as an
executioner caspase, is a classical marker of cell death by apoptosis. Caspase-3 is activated via
proteolytic cleavage at conserved aspartic residues, generating a large and a small subunit, which
consequently dimerize to form the active enzyme (Elmore, 2007). Therefore, caspase-3
activation is measured by cleaved caspase-3 protein, as detected by Western blotting. To assess
whether the isoforms can affect the activation of caspase-3 in response to cytokine-mediated
apoptosis, INS-1 cells were transduced with adenoviruses over-expressing PGC-1α1, and
PGC-1α4 or GFP as a control. 30 hours post-infection (allowing sufficient isoform expression),
cells were treated with a cytokine cocktail of TNFα, IL-1β and IFNγ, for 18 hours (to promote
cell death).
First, to confirm over-expression of the isoforms, a Western blot for PGC-1α was
performed (Figure 9A). This revealed that PGC-1α1 and PGC-1α4 were over-expressed to
similar levels and higher than the GFP-infected control. Interestingly, cytokine treatment reduces
protein expression of over-expressed PGC-1α1 and PGC-1α4; however, the underlying
mechanism is unknown. TNFα is known to decrease expression of PGC-1α in cardiac cells
53
(Palomer et al., 2008) and may be the cause for the decrease in PGC-1α1 and PGC-1α4 seen in
these blots. Another plausible explanation could be that the cytokines disrupt the viral or
CMV-promoted protein over-expression from the vector. Additionally, PGC-1α1 and PGC-1α4
seem to be induced by the expression of the other (Figure 9A). For this reason, it cannot be
determined whether the protective effect is from PGC-1α4 alone, or a combination of actions
from the two. To directly address this, over-expression would have to be induced with the
combination of short-hairpin RNA (shRNA) to prevent the endogenous induction of the opposite
isoform. Moreover, regardless of the mutual up-regulation of PGC-1α1 and PGC-1α4, protein
expression of each of the isoforms was more highly expressed as compared GFP, allowing a
comparison between the control and each isoform to be made in terms of impact on cell death
pathways.
To determine whether cytokines have an effect on the apoptotic profile of INS-1 cells, the
activation of caspase-3 was investigated by a Western blot for cleaved caspase-3, using the same
protein lysates used for the PGC-1α Western blot (Figure 9B). There is some degree of cleaved
caspase-3 present in control samples without cytokine treatment, probably due to cytotoxicity
from using virus for over-expression. However, in the presence of cytokines, there is a more
robust increase of caspase-3 cleavage when observing protein in the GFP control lanes. There
was an evident decrease of cleaved caspase-3 in the INS-1 cells over-expressing only PGC-1α4
in the presence of cytokines (Figure 9B and 9C). PGC-1α1 may also prevent caspase-3 cleavage,
to an intermediate degree; however, the decrease compared to the control was not statistically
significant (Figure 9C). Additionally, this moderate decrease may be due to the up-regulation of
endogenous PGC-1α4 in the sample. Furthermore, the basal levels of cleaved caspase-3
decreased in the presence of over-expressed PGC-1α1 and PGC-1α4 as compared to GFP, only
54
the difference between GFP and PGC-1α4 was statistically significant (Figure 9C); however,
there caspase-3 cleavage in the presence of PGC-1α1 is markedly decreased. These results
suggest that PGC-1α4, and possibly to a lesser degree PGC-1α1, may be anti-apoptotic. To
convincingly conclude whether PGC-1α1 and PGC-1α4 are anti-apoptotic in β-cells, other
experiments, such as protection from nucleosome fractionation, will need to be performed, since
active caspase-3 also has non-apoptotic functions (Aispuru et al., 2008; D'Amelio et al., 2011;
Fujita et al., 2008; McComb et al., 2010; Woo et al., 2003). What also remains unclear is how
these isoforms regulate this function at the molecular level, and if they do so in a similar or
unique manner.
PGC-1α1 and PGC-1α4 expression is regulated by GLP-1
The actions of PGC-1α4 are strikingly reminiscent to the anti-apoptotic effects of
glucagon-like peptide-1. Binding of GLP-1 to its receptor increases intracellular cAMP levels,
which has been shown to mediate GLP-1’s action on postprandial glucose-stimulated insulin
secretion. Interestingly, GLP-1 also has significant cyto-protective properties and can promote
β-cell survival in the presence of a cytotoxic chemical, Streptozotocin (STZ), ER-stress or
cytokine exposure (Cunha et al., 2009; Li et al., 2005; Riedel et al., 2010; Yusta et al., 2006).
GLP-1 regulates its cyto-protective actions by activation of cAMP, followed by activation of
CREB by phosphorylation. GLP-1 can decrease pro-apoptotic proteins, such as cleaved
caspase-3 and increase pro-survival factors such as Bcl-2 and Bcl-xL (Baggio and Drucker,
2007). However, how GLP-1 regulates its anti-apoptotic effects is still not well understood.
PGC-1α4 expression is induced by cAMP, which is downstream of GLP-1 signalling and
PGC-1α4 is capable of decreasing the activation of caspase-3 in the presence of cytokines,
55
similar to GLP-1. Therefore, we hypothesized that the anti-apoptotic actions of GLP-1 may be
mediated through activation of PGC-1α4.
The first step to determine whether PGC-1α4 mediates the actions of GLP-1 is to assess
whether GLP-1 receptor activation lies upstream of PGC-1α4 expression in β-cells. This was
accomplished by investigating the expression of PGC-1α1 and PGC-1α4 after treatment of
INS-1 cells with Exendin-4 (Ex-4), a long lasting GLP-1 analog. After 2 hours, expression of
PGC-1α4 mRNA was increased by ~6-fold and PGC-1α1 was moderately increased, by ~3.5-
fold (Figure 10A). Both PGC-1α1 and PGC-1α4 induction is clear at the mRNA level, however,
only PGC-1α4 expression is apparent after immunoblotting with an anti-PGC-1α antibody
(Figure 10B). PGC-1α4 protein is expressed ~3-fold higher in comparison to the untreated cells
(Figure 10C). The protein expression of PGC-1α4 goes from being undetectable in the control,
lanes to being expressed upon treatment with Ex-4. Since Ex-4 can increase expression of
PGC-1α4 in INS-1 cells, it is plausible that PGC-1α4 acts downstream of GLP-1 to mediate the
hormone’s anti-apoptotic effects.
Knock-down of PGC-1α isoforms, prevents streptozotocin induced hyperglycemia in vivo.
Since PGC-1a4 appears anti-apoptotic in vitro, we hypothesized that loss of PGC-1α
isoform expression would increase the susceptibility of β-cells to cytokine-mediated apoptosis.
To address this question, we turned to an in vivo β-cell specific knockout of PGC-1α and its
isoforms. PGC-1α floxed mice were crossed with mice carrying a MIP (mouse insulin
promoter)-CreER transgene, to allow for inducible, β-cell specific deletion of PGC-1α and its
isoforms, herein named PGC-1αfl/flCre+
. Both PCG-1αfl/flCre+
and age- and weight-matched
56
littermate controls, PGC-1αfl/fl
, were genotyped using a PCR strategy, and crossed until the
floxed allele was homozygous and mice were either positive or negative for MIP-CreER.
To understand how PGC-1α1 and PGC-1α4 regulate β-cell survival in a physiological
context of diabetes, a steptozotocin (STZ) model of diabetes was utilized. The experimental plan
was as described in Figure 11. Two weeks after PGC-1αfl/flCre+
and PGC-1αfl/fl
mice were
gavaged with tamoxifen, to induce ablation of all PGC-1α isoforms in pancreatic β-cells, they
were subjected to multiple low-dose streptozotocin (STZ) injections to induce hyperglycemia
and β-cell apoptosis. One group of mice (n=7-8) was sacrificed 1 day after the final STZ
injection to quantify the amount of activated caspase-3 occurring in the β-cells at this early time
point. This was to address the question of whether the loss of PGC-1α isoforms modulates the
levels of apoptosis in vivo. Since STZ also impacts β-cell function and is a model of type 1
diabetes, we also evaluated the impact of PGC-1α loss on glucose homeostasis, 11, 18 and 32
days post the last STZ injection, by performing an early oral glucose tolerance test (OGTT), a
fasted/refed experiment, and a late OGTT, respectively. Furthermore, to evaluate the effect of
STZ on the development of hyperglycemia, random blood glucose was taken bi-weekly after the
first injection of STZ.
Interestingly, PGC-1αfl/flCre+
mice were protected against STZ induced hyperglycemia
over the course of the experiment (Figure 12A). PGC-1αfl/flCre+
mice trended to have lower
glycemia as compared to the controls; however, only the last blood glucose measurement was
significantly different. Furthermore, the early OGTT revealed that PGC-1αfl/flCre+
mice, 11 days
after the final STZ injection, were more glucose tolerant than the controls (Figure 12B), as
determined by a 2way ANOVA on the glucose curves and area under the curve (Figure 12B).
However, 32 days after the last STZ injection, both PGC-1αfl/flCre+
and PGC-1αfl/fl
mice had the
57
same glucose tolerance (Figure 12E). Additionally, all test mice at this late time point had high
fasting glucose levels, and could not clear their glucose back to fasting levels. This could be
because maximal β-cell death had occurred in both groups after this extensive period of time, and
differences in their glucose response are too subtle to detect or are out of detection range by the
glucose meters used.
Furthermore, STZ treated PGC-1αfl/flCre+
mice tend to have lower glucose after an
overnight fast and 2 hours after re-feeding, in comparison to controls (Figure 12C). As well,
PGC-1αfl/flCre+
mice tend to have higher circulating insulin after fasting and re-feeding, partially
explaining their lower glucose levels (Figure 12D). These results, however, did not reach
statistical significance, due to high variability but are similar to results shown in hyperglycemic
mice following siRNA-mediated knock-down of islet PGC-1α (Kim et al., 2009).
The mechanism to explain the protection of PGC-1αfl/flCre+
mice from STZ induced
hyperglycemia remains elusive and could be unrelated to the extent of β-cell death, yet instead
related to effects of PGC-1α isoforms on regulation of β-cell metabolism or insulin secretion.
Thus, to address the role of PGC-1α1 and PGC-1α4 in β-cell apoptosis, and whether it
contributes to the protection against STZ induced hyperglycemia in this model, pancreas samples
from mice 1 day post-STZ injections were stained by IHC for cleaved caspase-3 (Figure 13A).
Since our hypothesis is that PGC-1α4 (and potentially PGC-1α1) is anti-apoptotic, we predicted
that the KO mice would have more activation of cleaved caspase-3. Unexpectedly, the there was
no difference in caspase-3 immunostaining between these two groups visually (Figure 13A).
This was confirmed by quantification of the percentage of positive cells for cleaved caspase-3
staining in islets (Figure 13B). Approximately 30 – 35% of cells in each islet were positive for
caspase-3 in both groups and there was no statistical difference between the two groups (Figure
58
13B). This suggests that the protection against STZ induced hyperglycemia is not due to a
difference in β-cell apoptosis and points to a difference in β-cell function. It also indicated that
loss of PGC-1α isoforms in β-cells does not impact cell survival in a streptozotocin model of
β-cell apoptosis.
There are many reports that expression of certain cre-recombinase transgenes alone in
cells can impact cell function dependent of target gene excision (Lee et al., 2005). Therefore it
was important to assess whether the MIP-CreER transgene itself was contributing at all to the
glycemic phenotype or β-cell death. As a control for potential effects of cre-transgene expression
alone, a group of WT (wild-type) and MIP-Cre alone mice (n=7) were gavaged tamoxifen and
given the same STZ dose as all experimental mice, following the same experimental plan as the
KO mice up until the STZ injections. These mice were sacrificed 1-day post-STZ injections to
determine whether the cre transgene influences the phenotype. Again, there was no difference in
cleavage of caspase-3 between the WT and MIP-Cre only mice either visually (Figure 14A) or
following quantification (Figure 14B). Approximately 25 – 30% of the cells in the islet stained
for caspase-3; however there was also no statistical difference between the WT and MIP-Cre
only controls (Figure 14B). This demonstrated that the presence of the cre transgene does not
impact STZ-induced β-cell apoptosis in our experiment.
59
SECTION IV:
DISCUSSION
60
Summary of Results
In this study, we set forth to determine whether PGC-1α isoforms are endogenously
expressed and regulated in pancreatic β-cells. This study is the first attempt to understand the
function of different PGC-1α isoforms in β-cells. We report the novel finding that PGC-1α1 and
PGC-1α4 are induced at the mRNA and protein level by cAMP signalling, a cytokine cocktail
and Exendin-4, in either INS-1 cells or primary murine islets. To understand the biological
consequence of isoform induction by cytokines, we investigated whether PGC-1α1 and PGC-1α4
play a role in cytokine-mediated apoptosis. We demonstrated that PGC-1α4 over-expression in
INS-1 cells prevented the activation of caspase-3 to a greater extent than PGC-1α1, suggesting
differential regulation of apoptosis. We also showed that PGC-1α4 expression is promoted by
Ex-4, a long lasting GLP-1 analog, at the mRNA and protein level, suggesting a potential for
PGC-1α4 to act downstream of GLP-1 action in β-cells. Finally, to understand the function of
β-cell expression of PGC-1α1 and PGC-1α4 within a physiological context in vivo, PGC-1α
isoforms were knocked out specifically in mouse β-cells and mice were subjected to low-dose
STZ to cause β-cell death by apoptosis. Unexpectedly, these mice were protected against STZ
mediated hyperglycemia and more glucose tolerant compared to the controls at an earlier
time-point. Immunohistochemistry of pancreatic sections from mice 1 day after the last STZ
injection revealed that there was no difference in activation of caspase-3 in control versus KO
mice. This suggests differences in β-cell function rather than apoptosis as a possible mechanism
for the protection against STZ induced hyperglycemia in β-cell specific PGC-1α KO mice.
61
Regulation and Function of PGC-1α Isoforms in β-cells
It is clear from this data that PGC-1α1 and PGC-1α4 are expressed and likely have a
significant physiological role in β-cells; however, expression of other published isoforms
remains inconclusive or untested at this time. Our focus was on the isoforms transcribed from the
alternative promoter, since PGC-1α4, was found to have a distinct function from PGC-1α1 in
muscle (Ruas et al., 2012). Although, PGC-1α-b, PGC-1α-c, PGC-1α2, and PGC-1α3
(Chinsomboon et al., 2009; Ruas et al., 2012) were not found to have unique functions in muscle,
we also investigated their expression in β-cells since they are also transcribed from the
alternative promoter and may have a unique function in β-cells.
PGC-1α2 / PGC-1α-b was not expressed in β-cells and while PGC-1α3 / PGC-1α-c
mRNA was detectable, its low mRNA expression level, and undetectable protein expression,
suggests that it might not play a significant role in β-cell function. However, to rule out the
possibility that the detection level of our assays is not sensitive enough to detect low, yet
biologically relevant expression and regulation, immunoprecipitation can be performed to enrich
and concentrate PGC-1α proteins.
In contrast to these isoforms, we demonstrated that PGC-1α1 and PGC-1α4 were induced
by forskolin (a cAMP activator), a cytokine cocktail, and the GLP-1 analog, exendin-4. With
forskolin treatment, induction of PGC-1α1 and PGC-1α4 is variable between INS-1 and islets,
since expression of the isoforms as determined by qPCR are more robust in INS-1 cells
compared to islets. This is a caveat of using immortalized cell lines versus primary cell culture.
Since primary isolated islets are more physiologically relevant, we expect that the induction of
62
PGC-1α1 and PGC-1α4 are better reflected in the primary islet, rather than INS-1. However,
both models support that PGC-1α1 and PGC-1α4 are expressed and inducible in β-cells.
Even though mRNA expression of both PGC-1α1 and PGC-1α4 was induced for all of
the stimuli studied, PGC-1α1 protein expression was detected only after forskolin treatment, and
was not visible following exposure of β-cells to cytokines or Ex-4. PGC-1α4 is known to be
more stable than PGC-1α1 (Ruas et al., 2012), since PGC-1α1 is rapidly degraded by the
proteosome (Zhang et al., 2009). PGC-1α1 can be stabilized by p38 phosphorylation,
downstream of cAMP signalling, on residues in its repression domain (Cao et al., 2001; Fan et
al., 2004). This suggests that treatment with forskolin may have stabilized the PGC-1α1 protein
in β-cells. Interestingly, cytokines (TNFα, IL-1β and IL-1α) can also stabilize PGC-1α in muscle
(Puigserver et al., 2001); however, our data suggest that cytokines and exendin-4 do not have the
same stabilizing effect on the PGC-1α1 protein in β-cells. Again, to exclude the possibility that
the protein is low to detect using available antibodies, an IP can be performed to enrich and
concentrate the protein or pulse chase experiments can be performed with radioactivity to
increase sensitivity of detection. Cells could also be treated with a proteosomal inhibitor MG132
(Zhang et al., 2009) to stabilize the protein, making it visual by Western blotting. In contrast, it is
clear PGC-1α4 mRNA and protein is abundant after treatment with cytokines or Exendin-4. Our
data suggests that PGC-1α4 may play a significant role in the intracellular response to these
stimuli, but at this time it cannot be determined whether PGC-1α1 plays a physiologically
significant function under these conditions.
Interestingly, unlike INS-1 cells, PGC-1α4 expression was undetectable by qPCR in
unstimulated islets. Only upon treatment with forskolin did PGC-1α4 mRNA become detectable.
Currently, we have only investigated induction of these isoforms with forskolin stimulation in
63
islets, and Ex-4 and cytokine experiments are planned to determine if a similar induction of
PGC-1α4 is observed in islets as with INS-1 cells. Regardless of differences in unstimulated
expression of PGC-1α4 between INS-1 and primary islets, PGC-1α4 was significantly induced at
the mRNA and protein level in both in vitro and ex vivo models by extracellular signals,
suggesting dynamic regulation of this molecule in response to external stimuli.
Increased levels of circulating pro-inflammatory cytokines, particularly TNFα, IL-1β and
IFNγ are hallmarks of both type 1 and type 2 diabetes (Donath et al., 2003; Fain, 2006). Chronic
stimulation by pro-inflammatory factors causes β-cells dysfunction and death. Since PGC-1α4
protein is induced after stimulation with a cytokine cocktail in vitro, this suggests that in vivo
PGC-1α4 may also be induced in β-cells during the progression of diabetes. Ultimately, this
could be tested by determining the relative expression of PGC-1α4 in islets from diabetic
patients. Additionally, our data suggests that PGC-1α4 may be anti-apoptotic through prevention
of caspase-3 activation, so PGC-1α4 could potentially be induced when β-cells are exposed to
chronic levels of circulating cytokines, as a mechanism to protect against β-cell. Moreover,
PGC-1α4 is induced by cAMP signalling and exendin-4, which promote β-cell survival (Jambal
et al., 2003; Jhala et al., 2003; Li et al., 2005; Li et al., 2003). These findings support our
hypothesis that PGC-1α4 is anti-apoptotic in β-cells, since the co-activator can act downstream
of these cyto-protective factors, and thus may, in part, mediate their anti-apoptotic effects. This is
therapeutically relevant as GLP-1 receptor agonists, such as Exenatide and Liraglutide, are
current pharmacological interventions to treat type 1 and type 2 diabetes (Lovshin and Drucker,
2009), but their complete mechanism of action on β-cells is not yet understood. Therefore,
PGC-1α4 induction by GLP-1 may be a mechanism that mediates or improves the effectiveness
of these drugs to treat disease through promotion of β-cell survival. To test this theory one could
64
knock down PGC-1α4 in INS-1 cells and islets and determine whether exendin-4 or cAMP still
promotes β-cell survival in response to cytotoxic signals such as pro-inflammatory cytokines in
the absence of co-activator expression. If the anti-apoptotic function of GLP-1 or cAMP is lost
when PGC-1α4 is knocked down, this would suggest that these factors are dependent on
PGC-1α4 expression to exert their pro-survival functions.
A caveat of our in vitro experiments is the efficiency of infection by the adenoviruses. To
achieve detectable levels of expression while maintaining health of the cultures, we could only
obtain an infection efficiency of 50-60%. Thus, cells that were not infected may have diluted out
the biological responses of the cells that we were measuring. Therefore, prevention of caspase-3
activation by PGC-1α4 could actually be a more robust response than we observed. In addition, it
was evident from our experiments that simple transduction of INS-1 cells with viral vectors
activated apoptotic pathways, as shown by the detection of cleaved caspase-3 in untreated
samples. It is possible that the virus primes the cells for cell death, amplifying or modulating the
cytoprotective effects of PGC-1α isoform over-expression. To eliminate this variable, stable lines
over-expressing PGC-1α1 and PGC-1α4 were generated and will be used to test this theory in
future experiments. PGC-1α1 and PGC-1α4 were cloned into the multiple cloning site of
pCDNA, downstream of a CMV promoter and with a N-terminal FLAG tag. INS-1 cells were
transfected with linear plasmid and following successful integration of the linearized plasmid,
cells were selected for using G418 antibiotic. These cells will be used to circumvent the use of
virus to over-express PGC-1α1 and PGC-1α4. Additionally, this model will ensure
approximately 100% of the cells are over-expressing their respective isoform and ameliorate the
poor infection efficiency of the viruses.
65
To confirm PGC-1α4’s anti-apoptotic function, more conclusive assays can be performed
that directly assess cell death, as cleaved caspase-3 is also known to have apoptosis-independent
functions. Caspase-3 has been shown to be a critical factor in; (i) erythroid cell expansion and
survival (Aispuru et al., 2008); (ii) CD8+ T-cell expansion (McComb et al., 2010); (iii) stem cell
differentiation (Fujita et al., 2008); (iv) inhibition of B-cell proliferation through loss of
repression of a cell cycle promoter, Cdkn1a (Woo et al., 2003) and; (v) dendrite synaptic
dysfunction (D'Amelio et al., 2011). Although these functions are very cell-type specific, it
confirms that activation of caspase-3 does not necessarily lead to an increase of apoptosis. So,
when studying caspase-3 as a marker for apoptosis, it is critical to confirm whether an increase in
caspase-3 activation truly corresponds to an increase in apoptosis. One such assay is a
nucleosome ELISA that measures the amount DNA fragmentation, which is indicative of cells
undergoing apoptosis. The effects of isoform over-expression on apoptosis could also be
determined with terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL)
staining, which labels DNA fragmentation (Gavrieli et al., 1992). Even if we are limited by
transduction or transfection efficiency, this assay allows us to determine the percentage of cells
positive for GFP (co-expressed with PGC-1α isoforms) and negative for TUNEL, revealing
which cells are protected against cell death. Another method could be to sort cells by flow
cytometry. Cells undergoing cell death by apoptosis are positive for Annexin V. If PGC-1α4
and/or PGC-1α1 protect INS-1 cells from apoptosis, GFP positive cells will be Annexin V
negative, which will indicate whether cells directly expressing the isoform are protected against
apoptosis. These experiments will help to convincingly determine whether over-expression of
PGC-1α4 in INS-1 cells protects cells against apoptosis and if this is a function unique from
PGC-1α1.
66
It has proven technically difficult to measure cleaved caspase-3 in primary islets;
however addressing our hypothesis in this model would better support an
anti-apoptotic role in vivo. It is substantially more difficult to transduce whole islets, compared to
INS-1 cells with adenovirus. The virus fails to penetrate the inner cells of the islet, most of which
are the β-cells. Additionally, it has been difficult to induce apoptosis in whole islets by
cytokines. These conditions are being optimized and different methods to infect islets (such as
using dispersed cells rather than whole islets) and conditions to induce apoptosis (such as high
concentrations of cytokines, STZ, fatty acids or a combination or all three) are being tested.
Therefore, our data suggests that PGC-1α4 has the potential to be anti-apoptotic in
β-cells, which is a novel function for this protein. However, further studies are needed to
elucidate the relevance of this function in vivo and within the context of β-cell disease.
Isoforms and the Pathogenesis of Diabetes
Since promoting β-cell survival is an attractive diabetes therapy, we were interested in the
role of PGC-1α isoforms within a physiological context. Using readily available tools, we first
chose to use a knockout of all PGC-1α isoforms. Since PGC-1α is important in metabolically
active tissues (Lin et al., 2005) and is a significant regulator of mitochondrial function, it is
surprising that β-cell specific PGC-1α KO mice were protected against STZ induced
hyperglycemia. However, our results are consistent with studies investigating the role of PGC-1α
in rodent β-cells. PGC-1α1 over-expression in mouse β-cells causes glucose intolerance and
decreases insulin secretion, whereas knock-down causes the inverse (De Souza et al., 2005; De
Souza et al., 2003; Kim et al., 2009; Valtat et al., 2013; Yoon et al., 2003). These results suggest
that constitutive PGC-1α1 expression may be detrimental to β-cell function. These findings
67
seemingly contradict that of data in human islets, since OXPHOS genes and PGC-1α expression
are decreased in islets from diabetic patients and knock-down of PGC-1α in human islets
decreases insulin secretion (Ling et al., 2008; Olsson et al., 2011). The techniques used to
manipulate PGC-1α gene expression in vivo and in vitro in rodent models, may contribute to the
inability of rodent models of diabetes to recapitulate the human etiology and underscores the
importance of repeating experiments with human tissues and/or cell lines. Additionally, these
studies fail to differentiate between potential differences in PGC-1α isoform function in β-cells
in the context of human disease, which adds complexity in understanding the role of PGC-1α in
the pathogenesis of diabetes. Our in vitro studies implicate PGC-1α4 as a protective factor in
β-cells and is therefore a potential target to prevent β-cell death in diabetic patients. However,
this effect must be further studied in vivo. Since the in vivo KO is of all isoforms, PGC-1α4’s
specific contribution to the phenotype remains unclear. Elucidating the effect of PGC-1α4 in
vivo, by rescue experiments, can confirm whether PGC-1α4 is also protective in a physiological
context. This will provide additional evidence to support targeting specific isoforms, namely
PGC-1α4, to improve β-cell function or survival.
To determine whether a change in the apoptotic status of β-cells is a contributing factor to
the glycemic phenotype of the beta-cell specific PGC-1α KO, mice were sacrificed soon after the
final STZ injection and pancreas sections were analyzed by IHC for a marker of apoptosis. IHC
analysis revealed that there was no difference in activation of caspase-3 between the knock-outs
and controls. There was also no difference in the amount of caspase-3 activation between control
mice expressing only the MIP-Cre transgene, confirming that there are no non-specific effects of
cre recombinase expression on caspase-3 activation. These data suggest that the protection from
STZ induced hyperglycemia in β-cell specific PGC-1α KO mice is due to a difference in β-cell
68
function rather than a change in the rate of cell death. However, since activation of caspase-3
does not absolutely determine whether a cell is undergoing apoptosis, an alternative method of
quantify apoptotic β-cell death is required, such as TUNEL or Annexin V staining in vivo, to
confirm these results.
A caveat to knocking out all PGC-1α isoforms is the difficulty in differentiating the
contribution of each isoform to the phenotype. To generate a floxed mouse where PGC-1α4 is
individually knocked out in β-cells is difficult because targeting any portion of PGC-1α4 will
also affect the expression of other isoforms or full length PGC-1α. Therefore, to understand the
role of PGC-1α1 and PGC-1α4 individually in an in vivo context, these isoforms could be
rescued separately in our beta-cell specific PGC-1α KO mice prior to STZ injections. This could
be achieved by utilizing AAV8 (adeno-associated virus 8)-MIP adenovirus (Rehman et al.,
2008). AAV8 is a viral vector that specifically targets and infects all cells within the pancreas.
The inclusion of the mouse insulin promoter in the vector restricts the over-expression of a gene
of interest to the β-cell (Rehman et al., 2005; Wang et al., 2006). A double stranded AAV vector
carrying PGC-1α1 or PGC-1α4 cDNA under the regulation of the mouse insulin promoter could
be generated to rescue PGC-1α1 and PGC-1α4 in vivo (in the absence of other isoforms) in order
to discern their individual roles in rodent models of diabetes.
Streptozotocin as a model of type 1 diabetes has limitations. It is a pharmacological toxin
and inducer of β-cell death, but may also affect any other cell that expresses a functioning Glut2
receptor. In addition, it is not known whether the mechanism by which β-cells die in response to
STZ is similar to the β-cell apoptosis observed in type 1 diabetic patients. To further delve into
the role of PGC-1α4 (and other isoforms) in the pathogenesis of type 1 diabetes, studies with the
NOD mouse may be more informative and address an alternative mechanism of β-cell mediated
69
death in vivo and will complement the studies using streptozotocin. To determine whether there
are protective functions of PGC-1α1 and PGC-1α4 against β-cell death in an alternative model of
diabetes that better mimics the disease pathogenesis, NOD mice can also be injected with the
AAV8 viruses over-expressing PGC-1α1 and PGC-1α4 after birth and after the disease has
progressed. This will investigate whether direct delivery of these isoforms to the β-cells of mice
before and after the development of diabetes can prevent the progression of the disease and
β-cell in vivo. This model can provide more direct evidence that PGC-1α4 is capable of
preventing the pathogenesis of diabetes and whether this is a unique function from PGC-1α1 in
β-cells. The NOD mouse develops diabetes by immune infiltration and since genes regulated by
PGC-1α4 are revealed upon TNFα signalling in primary hepatocytes (unpublished), the NOD
mouse may be a more direct in vivo model of PGC-1α4 because of the pro-inflammatory
environment.
Even though there is an apparent disconnect between human and rodent data, studying
mouse islets is still relevant to understand the underlying mechanisms of PGC-1α’s involvement
in diabetes. Mouse islets are used tools to understand complex processes in humans. Although
data from mice may not identically recapitulate the human condition, they are useful to elucidate
mechanisms since human tissue is difficult to obtain. Furthermore, discrepancies between
rodents and humans may be related to the methods chosen to manipulate gene expression. For
example, in the rodent studies investigating PGC-1α function in β-cells, PGC-1α knock-down
includes all isoforms, whereas over-expression is restricted to PGC-1α1, which could explain the
differences in rodent and human data. Additionally, there have not been any studies that
genetically knock-down only PGC-1α1 in vivo, which will more closely mimic the human
condition. Furthermore, the studies do not address the potential contribution of other PGC-1α
70
isoforms in diabetes, since their functions are not very well understood in β-cells. Therefore, as
this project develops it is essential to evaluate the role of PGC-1α isoforms, particularly the
differences between PGC-1α1 and PGC-1α4 in human tissues. Since these isoforms may play a
critical role in rodent β-cells, it will be important to determine whether these isoforms are also
important in human β-cells. As the function of PGC-1α4 is elucidated in β-cells, investigating
whether these functions are conserved between different species will be relevant to human
disease.
Differential Role of PGC-1α1 and PGC-1α4 in β-cells
Published data demonstrates that PGC-1α1 and PGC-1α4 have unique functions in
muscle and PGC-1α4 does not regulate classical PGC-1α1 genes (Ruas et al., 2012). Our results
also suggest that PGC-1α1 and PGC-1α4 have differential functions in β-cells in terms of
apoptosis. We determined that PGC-1α4 is more effective at preventing the activation of
caspase-3, suggesting that it is a more effective anti-apoptotic factor than PGC-1α1. However,
this may be confounded by mutual up-regulation of each of the isoforms when either is
over-expressed (Figure 9A). Moreover, although we only see increased PGC-1α4 protein
following cytokine treatment, it cannot be confirmed whether PGC-1α4 is the only isoform
induced and regulated by cytokines and Ex-4 in INS-1 cells, since the presence of PGC-1α1
protein expression under these conditions is inconclusive due to limitations of the antibody and
protein instability.
Unpublished data from our lab in hepatocytes (a cell type that co-ordinately controls
whole body metabolism with β-cells and are derived from similar lineages), demonstrate that
differential gene expression between PGC-1α1 and PGC-1α4 is revealed in the presence of an
71
inflammatory stimulus, TNFα. We expect that the induction of PGC-1α1 and PGC-1α4 by
cytokines in β-cells also results in differential gene regulation and function in β-cells. Since this
is only speculation, it will be important to characterize differential gene expression patterns in a
β-cell line or primary islets. To do this, INS-1 cells and isolated islets can be transduced with
adenovirus expressing each isoform and treated with cytokines to investigate different
inflammatory pathways by qPCR of candidate mediators. However, this is a biased approach. An
unbiased approach would be to transduce INS-1 cells or islets with the different isoforms and
subject them to microarray analysis (or second generation genomics/deep sequencing), to reveal
all genes differentially regulated between PGC-1α1 and PGC-1α4. This would be a preliminary
study to determine whether different pathways are regulated in response to an inflammatory
stimulus, thus revealing potential differences in functional output between PGC-1α1 and PGC-
1α4 in β-cells. This experiment could also indicate the molecular pathways regulating the cyto-
protective properties of PGC-1α4.
PGC-1α4 versus NT-PGC-1α
Since both NT and PGC-1α4 are similarly spliced (Figure 4), have almost identical
sequences (Figure 5), and are approximately the same molecular weight (Ruas et al., 2012;
Zhang et al, 2009), it is difficult to differentiate these two proteins at the mRNA level with qPCR
primers or by Western blotting. The only difference between these isoforms is that NT-PGC-1α
is transcribed from the proximal promoter and PGC-1α4 is transcribed from the alternative
promoter, resulting in different exon 1 sequences. Primers to the alternative promoter and the
shared exon 1 can differentiate between NT-PGC-1α and PGC-1α4, but also recognize PGC-1α2.
Similarly, primers to the proximal promoter and the shared exon 1 can also differentiate between
72
NT-PGC-1α and PGC-1α4, but also recognize PGC-1α1. In addition, the available antibody
recognizes all known isoforms of PGC-1α and cannot differentiate between NT-PGC-1α and
PGC-1α4 because of their similar molecular weight. The similarity between these two isoforms
potentially means that the mRNA and protein expression we observe in our experiments is
actually NT-PGC-1α and not PGC-1α4.
An alternative means of differentiating between NT-PGC-1α and PGC-1α4 expression in
β-cells is to clone the full-length mRNA sequences by PCR. Using cDNA generated from INS-1
cells or isolated islets and primers specific for each isoform, full-length products can be
amplified. These PCR products would be subjected to sequencing analysis to confirm the identity
of the amplicons and relative expression levels compared by qPCR. This method will determine
the absolute expression of PGC-1α4 and NT-PGC-1α and demonstrate whether either or both
isoforms are expressed and inducible in β-cells. Again it will be critical to also determine the
relative abundance of both PGC-1α4 and NT-PGC-1α in human tissues. Although both are
expressed in many human tissues (Ruas et al., 2012; Zhang et al., 2009) some tissues appear to
have preference for one isoform over another and it is unknown which are expressed in human
islets. Additionally, even if they are both expressed in human β-cells they may not be similarly
regulated. Using rodent muscle as an example, over-expression of PGC-1α4 and NT-PGC-1α via
electroporation, allows accumulation of over-expressed PGC-1α4, but not NT-PGC-1α, protein
(Ruas et al., 2012). So even though these two proteins are similar, the distinct N-terminus of
PGC-1α4 is enough to cause differential regulation of these isoforms in muscle. Furthermore,
although they have very similar protein structures, only NT-PGC-1α regulates aspects of
mitochondrial function and does not have the myotrophic effects of PGC-1α4 (Ruas et al., 2012),
demonstrating that determining the precise identity of the protein in islets will be important when
73
trying to elucidate function. Therefore, it is plausible that PGC-1α4 and NT-PGC-1α may be
differentially regulated in human islets, as well as mouse islets.
Perspectives and Conclusions
PGC-1α4 is potentially a novel factor important for β-cell survival in the context of
diabetes. Pro-inflammatory cytokines induce expression of this protein, which could be a
mechanism in vivo to prevent cytokine induced β-cell death in diabetes. We have identified a
cyto-protective function for PGC-1α4 in β-cells; however, since PGC-1α4 is expressed in
multiple tissues, this function may be shared in other cells types. Since we only investigated this
one physiological outcome of PGC-1α4 biology, it is unclear whether PGC-1α4 has additional
unique function from PGC-1α1 in β-cells and whether they regulate different gene pathways. We
can speculate that this may be the case, because PGC-1α1 and PGC-1α4 regulate different gene
sets in muscle (Ruas et al., 2012) and in hepatocytes (our unpublished data). If these isoforms do
have different functions in β-cells, we would expect that even if they were similarly regulated by
forskolin or cytokines, for example, their functional output could be unique. A microarray
comparing mRNA expression patterns in PGC-1α1 and PGC-1α4 expressing INS-1 cells or islets
will direct us towards more potential functional differences between PGC-1α1 and PGC-1α4.
Since β-cells are highly dependent on mitochondria and PGC-1α1 is a master regulator of
mitochondrial function, it will be important to determine whether PGC-1α4 also has important
roles in mitochondrial functions in β-cells. It is probable that PGC-1α4 will not regulate
mitochondrial genes classically regulated by PGC-1α1 in β-cells, as this is the case in muscle
(Ruas et al., 2012). However, it is still important to determine whether this is tissue specific or a
general characteristic of PGC-1α4. This information can also be gleaned from microarray data
74
following over-expression of PGC-1α1 or PGC-1α4 in INS-1 cells or primary islets or classical
analysis of using qPCR, Western blot analysis, and ETC activity assays.
Although up-regulation of PGC-1α in obese mouse models is associated with β-cell
dysfunction (Yoon et al., 2003), this was limited to in vitro adenoviral over-expression of
PGC-1α in isolated islets and may not reflect accurately the consequence of PGC-1α endogenous
up-regulation in vivo. Therefore, PGC-1α induction in obese mice may be a reason why they do
not develop frank diabetes and β-cell apoptosis. It is possible that both PGC-1α1 and PGC-1α4
can contribute to this phenomenon by increasing metabolism and protecting against cell death in
β-cells, respectively. This would complement human studies, which demonstrate that
knock-down of PGC-1α decreases β-cell function (Ling et al., 2008), rather than increasing
β-cell function (De Souza et al., 2003; De Souza et al., 2005; Kim et al., 2009).
Since there was no difference in caspase-3 activation in the pancreatic sections of
PGC-1α knock-outs and littermate controls, this suggests that the modest protection from STZ-
induced hyperglycemia observed in β-cell specific PGC-1α KO mice is due to β-cell dysfunction
rather than a difference in apoptosis. It is surprising that a knock-out of PGC-1α caused an
increase in insulin secretion, since mitochondrial function is necessary for insulin secretion.
Yoon et al., attributed this to an imbalance in metabolic gene expression, such as inappropriately
increased glucose-6-phosphatase and glucokinase expression (Yoon et al., 2003). Likely the
change in insulin secretion is due to an effect of PGC-1α1, as PGC-1α4 has not been associated
with mitochondrial function (Ruas et al., 2012). However, investigating whether the rescue of
PGC-1α4 with AAV8 adenoviruses in vivo can alter mitochondrial function or affect insulin
secretion would address this question. If PGC-1α4 does affect mitochondrial function or genes,
75
this would be a novel function of PGC-1α4 in β-cells and it would demonstrate that
PGC-1α4 has unique tissue specific properties.
PGC-1α4’s function in β-cells appears to be a pro-survival factor. Thus, PGC-1α4’s
potential anti-apoptotic effects could be beneficial for the treatment of diabetes. Type 1 diabetes
is caused by autoimmune attack of the β-cells. The simplest and most common method of
treating T1D is by insulin injections to compensate for the loss of β-cell mass. In the past decade,
islet transplantation has become a more attractive option, since patients can become insulin
independent for 1 – 3 years (Bellin et al., 2008; Shapiro et al., 2000). However, islets that are
transplanted into patients eventually succumb to autoimmune attack resulting in β-cell death,
even if immunosuppressant drugs are administered (Monti et al., 2008). To improve islet
transplantation therapy, it is critical to identify factors that can improve the health and survival of
the transplanted islets to increase the efficacy of this treatment option. PGC-1α4’s potential anti-
apoptotic effects could be beneficial in this regard, as expression may improve the efficacy of
islet transplantation by protecting islets from cell death caused by immune-mediated attack. The
benefits of preventing β-cell death are not limited to T1D. T2D in later stages is also associated
with high levels of circulating pro-inflammatory cytokines and immune-mediated β-cell death.
We also show that both PGC-1α1 and PGC-1α4 are downstream of GLP-1 receptor signalling in
β-cells, implicating these isoforms in the actions of GLP-1 (which are generally beneficial).
Determining the role of each isoform in GLP-1 action and, in particular, whether PGC-1α4
mediates the cyto-protective properties of GLP-1 could improve the understanding of GLP-1
action in β-cells and be applied to improve the efficacy of GLP-1R agonist drugs.
Therefore, we showed that PGC-1α isoforms are expressed and inducible in β-cells,
particularly PGC-1α1 and PGC-1α4. PGC-1α4 may be anti-apoptotic in vitro and this may be
76
applicable to β-cell survival in vivo. If PGC-1α4 is found to prevent the pathogenesis of diabetes
in vivo, it is potentially a novel diabetic treatment either as a factor to protect β-cell death in islet
transplantation or to improve current pharmaceutical interventions.
77
SECTION V:
FIGURES
78
Figure 1. Summary of extrinsic and intrinsic pathways of apoptosis. The extrinsic pathway
works through binding of FasL and TNFα, to their cognate receptors. This action recruits adaptor
proteins which then activates caspase-8 by proteolytic cleavage. This initiator caspase can then
activate caspase-3, the executioner, to cause apoptosis. External stresses activate the intrinsic
pathway causing permeability changes of the the mitochondrial membrane. This allows proteins
such as cytochrome C and SMAC/DIABLO to be released by activation of caspase-3 or inhibition
of IAP, respectively. Anti-apoptotic Bcl-2 family members, including Bcl-2 itself and Bcl-XL
inhibit mitochondrial membrane permeabilization, effectively preventing the release of CytoC.
Pro-apoptotic Bcl-2 members, Bim, Bax and Bid, can either inhibit Bcl-2 and Bcl-XL, or promote
membrane permeabilization, causing cytochrome C release. Both the extrinsic and intrinsic
pathways converge on caspase-3 activation to commit the cell to apoptosis. (Adapted from:
Elmore, 2007)
79
Figure 2: Anti-apoptotic actions of GLP-1. GLP-1 can promote cell survival either
through adenylyl cyclase – CREB axis, or PI-3K – AKT axis. These pathways either
prevent apoptosis by activating the expression of Bcl-2 and Bcl-XL, or preventing the
activation of caspases. (Adapted from: Baggio and Drucker, 2007)
80
Figure 3. Functional domains of PGC-1α. PGC-1α has the following domains: activation,
transcriptional repression, arginine and serine (RS) rich domains, and RNA binding domain.
These domains can bind the HAT complex (in the activation domain) and the mediator
complex (in the RS and RNA binding domain). LXXLL are nuclear receptor binding sites.
Adapted from Lin et al., 2005.
81
Figure 4. Summary of PGC-1α isoform transcripts. Each colour represents different
exons as labelled for PGC-1α1. The canonical exon 1 is spliced out of PGC-1α-b and
PGC-1α-c. Alternative splicing of intro 6 in NT-PGC-1α and PGC-1α4 introduces an in
frame stop codon. PGC-1α2, PGC-1α3 and PGC-1α4 are transcribed from the alternative
promoter approximately 14 kb upstream of the canonical exon 1. Out of the three, PGC-
1α3 has a unique exon 1.
82
Figure 5. Nucleotide and protein sequence similarities between NT-PGC-1α and
PGC-1α4. Only the nucleotide sequence of exon 1 is unique between (A) NT-PGC-1α
and (C) PGC-1α4, highlighted in green. This corresponds to different N-terminal ends
of the protein sequence, seen highlighted in blue in (B) and (D).
83
Figure 6. Genotyping of mouse tail crude DNA. (A) Floxed and WT alleles were
genotyped using primers flanking the 5’ loxP site, generating the upper band for alleles with
the loxP site and a lower band for the WT allele. (B) MIP-Cre positive mice were genotyped
using a forward primer in the mouse insulin promoter and reverse primer in the cre
recombinase transgene. The combination of fl/fl alleles and Cre+ are PGC-1α knockout mice
and fl/fl alleles and Cre- were used as their littermate controls.
84
Figure 7. Endogenous expression of isoforms in a β-cell line and primary murine islets.
Cells were treated with either DMSO (vehicle) or 10μM Forskolin for 2 hours for mRNA
quantification and 4 hours for samples processed for total protein. (A) Analysis of gene
expression for isoforms expressed in INS-1 cells and (B) isolated murine islets. Gene
expression was analyzed by quantitative real-time PCR by using primers specific for
individual isoforms. Protein lysates from (C) INS-1 cells and (D) isolated islets were
immunoblotted for PGC-1α, using an antibody which recognizes all isoforms. Western blots
for (E) INS-1 and (F) islet PGC-1α isoform expression were quantified using ImageJ
software. Bars depict mean values, and error bars represent SD. *p<0.05 between indicated
group and control. ns = not significant.
85
Figure 8. Incubation with a cytokine cocktail regulates the expression of PGC-1α1 and
PGC-1α4 in a time-dependent manner. INS-1 cells were incubated without a cocktail of
cytokines, TNFα (50ng/ml), IFNγ (50ng/ml) and IL-1β (10ng/ml), for 6, 12 and 24 hours. (A)
Gene expression was analyzed by quantitative real-time PCR by using primers specific for
indicated genes. (B) Total protein was harvested from INS-1 cells treated with or without a
cytokine cocktail for 12 hours. Immunoblot of PGC-1α reveals PGC-1α4 expression in the cells
treated with cytokines. (C) Western blot quantification of PGC-1α4 using ImageJ software. Bars
depict mean values, and error bars represent SD. *p<0.05 between indicated group and control.
86
Figure 9. Overexpression of PGC-1α1 and PGC-1α4 reduces cleavage of caspase-3. INS-1 cells
were infected with adenovirus expressing PGC-1α1, PGC-1α4 and GFP as a control. 30 hours post-
infection, cells were treated with or without a cocktail of cytokines, TNFα (50ng/ml), IFNγ (50ng/ml)
and IL-1β (10ng/ml). After an 18 hour treatment with cytokines, cells were processed for total protein.
(A) Immunoblot for PGC-1α show overexpression PGC-1α1 and PGC-1α4. (B) Immunoblot with the
same protein lysates for cleaved caspase-3. Presence of PGC-1α4, and PGC-1α1 to a lesser extent,
reduces the amount of cleaved caspase-3. (C) Quantification of western blot in (B) using ImageJ. Bars
depict mean values, and error bars represent SD. *p<0.05 between indicated group and control.
87
Figure 10. Treatment of INS-1 cells with Exendin-4 induces the expression of PGC-1α1 and
PGC-1α4. INS-1 cells were treated with exendin-4 for 2 hours for mRNA quantification and 4
hours for total protein isolation. (A) Gene expression analysis was performed using qPCR for
indicated genes. Exendin-4 induces expression of PGC-1α1 and PGC-1α4. (B) Immunoblot for
PGC-1α shows an induction of PGC-1α4 protein. (C) Western blot quantification of PGC-1α4
using ImageJ software. Bars depict mean values, and error bars represent SD. *p<0.05 between
indicated group and control.
88
Figure 11. Experimental plan of PGC-1α KO with low-dose STZ injections. (A) Mice
were gavaged 100mg/kg tamoxifen for 10 days. After 2 weeks, mice were injected with
50mg/kg of STZ. Blood glucose was taken bi-weekly after the first STZ injection. Two groups
of mice (n=3) were sacrificed 1- and 6-days post STZ injections for histology of the pancreas.
Then, 11, 18 and 32 days post STZ injections, an early OGTT (n=9 PGC-1αflfl
, n=8 PGC-
1αflflCre+
), fasted/refed (n=9 PGC-1αflfl
, n=8 PGC-1αflflCre+
), and late OGTT (n=6 PGC-1αflfl
,
n=5 PGC-1αflflCre+
) were performed, respectively.
89
Figure 12. Knockout of PGC-1α and its isoforms in vivo. (A) Random fed glucose measurements of
PGC-1α knockout mice, n=10 compared to littermate controls, n=11, starting from the first day of
streptozotocin injections until the end of the experiment . (B) Early oral glucose tolerance (OGTT) test
between PGC-1α knock out mice and littermate controls (n=8 and n=9, respectively). Significance by
2way ANOVA. OGTT is quantified by area under the curve. (C) Fasted/refed experiment (n=5). (D)
Serum insulin content from fasted/refed experiment in (C). (D) Late OGTT between PGC-1α knock out
mice and littermate controls (n=6 and n=5, respectively). Data for (A) and (B) are representative of 2
independent experiments. Bars and points on the curve depict mean values, and error bars represent
SEM. *p<0.05 between indicated group and control, unless otherwise stated.
90
Figure 13. Cleaved caspase-3 expression in islets 1 day post-STZ injections of WT vs. KO
mice. (A) Visually, cleaved caspase-3 staining by immunohistochemistry was not different
between PGC-1αfl/fl
(n=8) and PGC-1αfl/flCre+
(n=7) groups. (B) Cleaved caspase-3 positive
cells were not statistically different between the groups when quantified. Error bars represent
SD. 10x magnification.
91
Figure 14. Cleaved caspase-3 expression in islets 1 day post-STZ injections of WT vs
MIP-Cre control mice. (A) Visually, cleaved caspase-3 staining by immunohistochemistry
was not different between WT (n=7) and MIP-Cre only controls (n=7). (B) Cleaved caspase-3
positive cells were not statistically different between the groups when quantified. Error bars
represent SD. 10x magnification.
92
SECTION VI:
TABLES
93
Table 1: List of primers used for PCR
Name Sequence (5’-- 3’)
Alpha floxed
For: TCCAGTAGGCAGAGATTTATGAC
Rev: TGTCTGGTTTGACAATCTGCTAGGTC
MIP-Cre
For: TAAGGGCCCAGCTATCAATGGGAA
Rev: GTGAAACAGCATTGCTGTCACTT
Table 2: List of primers used for qPCR
Name Sequence (5’-- 3’)
PGC-1α1
For: GGACATGTGCAGCCAAGACTCT
Rev: CACTTCAATCCACCCAGAAAGCT
PGC-1α2
For: CCACCAGAATGAGTGACATGGA
Rev: GTTCAGCAAGATCTGGGCAAA
PGC-1α3
For: AAGTGAGTAACCGGAGGCATTC
Rev: TTCAGGAAGATCTGGGCAAAGA
PGC-1α4
For: TCACACCAAACCCACAGAAA
Rev: CTGGAAGATATGGCACAT
MCP-1
For: TGCTGTCTCAGCCAGATGCAGTTA
Rev: TACAGCTTCTTTGGGACACCTGCT
94
References
Aispuru, G.R., Aguirre, M.V., Aquino-Esperanza, J.A., Lettieri, C.N., Juaristi, J.A., and
Brandan, N.C. (2008). Erythroid expansion and survival in response to acute anemia stress: the
role of EPO receptor, GATA-1, Bcl-xL and caspase-3. Cell biology international 32, 966-978.
Alnemri, E.S., Livingston, D.J., Nicholson, D.W., Salvesen, G., Thornberry, N.A., Wong, W.W.,
and Yuan, J. (1996). Human ICE/CED-3 protease nomenclature. Cell 87, 171.
Andersson, U., and Scarpulla, R.C. (2001). Pgc-1-related coactivator, a novel, serum-inducible
coactivator of nuclear respiratory factor 1-dependent transcription in mammalian cells.
Molecular and cellular biology 21, 3738-3749.
Arany, Z., He, H., Lin, J., Hoyer, K., Handschin, C., Toka, O., Ahmad, F., Matsui, T., Chin, S.,
Wu, P.H., et al. (2005). Transcriptional coactivator PGC-1 alpha controls the energy state and
contractile function of cardiac muscle. Cell metabolism 1, 259-271.
Baar, K., Wende, A.R., Jones, T.E., Marison, M., Nolte, L.A., Chen, M., Kelly, D.P., and
Holloszy, J.O. (2002). Adaptations of skeletal muscle to exercise: rapid increase in the
transcriptional coactivator PGC-1. FASEB journal : official publication of the Federation of
American Societies for Experimental Biology 16, 1879-1886.
Baggio, L.L., and Drucker, D.J. (2007). Biology of incretins: GLP-1 and GIP. Gastroenterology
132, 2131-2157.
Barroso, I., Luan, J., Sandhu, M.S., Franks, P.W., Crowley, V., Schafer, A.J., O'Rahilly, S., and
Wareham, N.J. (2006). Meta-analysis of the Gly482Ser variant in PPARGC1A in type 2 diabetes
and related phenotypes. Diabetologia 49, 501-505.
Bell, G.I., Horita, S., and Karam, J.H. (1984). A polymorphic locus near the human insulin gene
is associated with insulin-dependent diabetes mellitus. Diabetes 33, 176-183.
Bellin, M.D., Kandaswamy, R., Parkey, J., Zhang, H.J., Liu, B., Ihm, S.H., Ansite, J.D., Witson,
J., Bansal-Pakala, P., Balamurugan, A.N., et al. (2008). Prolonged insulin independence after
islet allotransplants in recipients with type 1 diabetes. American journal of transplantation :
official journal of the American Society of Transplantation and the American Society of
Transplant Surgeons 8, 2463-2470.
Brownlee, M. (2005). The pathobiology of diabetic complications: a unifying mechanism.
Diabetes 54, 1615-1625.
Butler, A.E., Janson, J., Bonner-Weir, S., Ritzel, R., Rizza, R.A., and Butler, P.C. (2003). Beta-
cell deficit and increased beta-cell apoptosis in humans with type 2 diabetes. Diabetes 52, 102-
110.
Cao, W., Medvedev, A.V., Daniel, K.W., and Collins, S. (2001). beta-Adrenergic activation of
p38 MAP kinase in adipocytes: cAMP induction of the uncoupling protein 1 (UCP1) gene
requires p38 MAP kinase. The Journal of biological chemistry 276, 27077-27082.
95
Chen, C., Hosokawa, H., Bumbalo, L.M., and Leahy, J.L. (1994). Mechanism of compensatory
hyperinsulinemia in normoglycemic insulin-resistant spontaneously hypertensive rats.
Augmented enzymatic activity of glucokinase in beta-cells. The Journal of clinical investigation
94, 399-404.
Chinsomboon, J., Ruas, J., Gupta, R.K., Thom, R., Shoag, J., Rowe, G.C., Sawada, N.,
Raghuram, S., and Arany, Z. (2009). The transcriptional coactivator PGC-1alpha mediates
exercise-induced angiogenesis in skeletal muscle. Proceedings of the National Academy of
Sciences of the United States of America 106, 21401-21406.
Collombat, P., Xu, X., Heimberg, H., and Mansouri, A. (2010). Pancreatic beta-cells: from
generation to regeneration. Semin Cell Dev Biol 21, 838-844.
Cory, S., and Adams, J.M. (2002). The Bcl2 family: regulators of the cellular life-or-death
switch. Nat Rev Cancer 2, 647-656.
Cowell, R.M., Blake, K.R., Inoue, T., and Russell, J.W. (2008). Regulation of PGC-1alpha and
PGC-1alpha-responsive genes with forskolin-induced Schwann cell differentiation. Neuroscience
letters 439, 269-274.
Cunha, D.A., Ladriere, L., Ortis, F., Igoillo-Esteve, M., Gurzov, E.N., Lupi, R., Marchetti, P.,
Eizirik, D.L., and Cnop, M. (2009). Glucagon-like peptide-1 agonists protect pancreatic beta-
cells from lipotoxic endoplasmic reticulum stress through upregulation of BiP and JunB.
Diabetes 58, 2851-2862.
D'Amelio, M., Cavallucci, V., Middei, S., Marchetti, C., Pacioni, S., Ferri, A., Diamantini, A.,
De Zio, D., Carrara, P., Battistini, L., et al. (2011). Caspase-3 triggers early synaptic dysfunction
in a mouse model of Alzheimer's disease. Nature neuroscience 14, 69-76.
D’Errico, I., Salvatore, L., Murzilli, S., Lo Sasso, G., Latorre, D., Martelli, N., Egorova, A.V.,
Polishuck, R., Madeyski-Bengtson, K., Lelliott, C., et al. (2011). Peroxisome proliferator-
activated receptor-gamma coactivator 1-alpha (PGC1alpha) is a metabolic regulator of intestinal
epithelial cell fate. Proc Natl Acad Sci 108, 6603-6608.
Danaei, G., Finucane, M.M., Lu, Y., Singh, G.M., Cowan, M.J., Paciorek, C.J., Lin, J.K.,
Farzadfar, F., Khang, Y.H., Stevens, G.A., et al. (2011). National, regional, and global trends in
fasting plasma glucose and diabetes prevalence since 1980: systematic analysis of health
examination surveys and epidemiological studies with 370 country-years and 2.7 million
participants. Lancet 378, 31-40.
De Souza, C.T., Araujo, E.P., Prada, P.O., Saad, M.J., Boschero, A.C., and Velloso, L.A. (2005).
Short-term inhibition of peroxisome proliferator-activated receptor-gamma coactivator-1alpha
expression reverses diet-induced diabetes mellitus and hepatic steatosis in mice. Diabetologia 48,
1860-1871.
96
De Souza, C.T., Gasparetti, A.L., Pereira-da-Silva, M., Araujo, E.P., Carvalheira, J.B., Saad,
M.J., Boschero, A.C., Carneiro, E.M., and Velloso, L.A. (2003). Peroxisome proliferator-
activated receptor gamma coactivator-1-dependent uncoupling protein-2 expression in pancreatic
islets of rats: a novel pathway for neural control of insulin secretion. Diabetologia 46, 1522-
1531.
Deacon, C.F., Nauck, M.A., Toft-Nielsen, M., Pridal, L., Willms, B., and Holst, J.J. (1995). Both
subcutaneously and intravenously administered glucagon-like peptide I are rapidly degraded
from the NH2-terminus in type II diabetic patients and in healthy subjects. Diabetes 44, 1126-
1131.
Degn, K.B., Juhl, C.B., Sturis, J., Jakobsen, G., Brock, B., Chandramouli, V., Rungby, J.,
Landau, B.R., and Schmitz, O. (2004). One week's treatment with the long-acting glucagon-like
peptide 1 derivative liraglutide (NN2211) markedly improves 24-h glycemia and alpha- and
beta-cell function and reduces endogenous glucose release in patients with type 2 diabetes.
Diabetes 53, 1187-1194.
Del Prato, S., and Marchetti, P. (2004). Beta- and alpha-cell dysfunction in type 2 diabetes.
Horm Metab Res 36, 775-781.
Donath, M.Y., Storling, J., Maedler, K., and Mandrup-Poulsen, T. (2003). Inflammatory
mediators and islet beta-cell failure: a link between type 1 and type 2 diabetes. Journal of
molecular medicine (Berlin, Germany) 81, 455-470.
Dor, Y., Brown, J., Martinez, O.I., and Melton, D.A. (2004). Adult pancreatic beta-cells are
formed by self-duplication rather than stem-cell differentiation. Nature 429, 41-46.
Du, C., Fang, M., Li, Y., Li, L., and Wang, X. (2000). Smac, a mitochondrial protein that
promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102,
33-42.
Ek, J., Andersen, G., Urhammer, S.A., Gaede, P.H., Drivsholm, T., Borch-Johnsen, K., Hansen,
T., and Pedersen, O. (2001). Mutation analysis of peroxisome proliferator-activated receptor-
gamma coactivator-1 (PGC-1) and relationships of identified amino acid polymorphisms to Type
II diabetes mellitus. Diabetologia 44, 2220-2226.
Elayat, A.A., el-Naggar, M.M., Tahir, M. (1995). An immunocytochemical and morphometric
study of the rat pancreatic islets. J Anat 186, 629-637.
Elmore, S. (2007). Apoptosis: a review of programmed cell death. Toxicologic pathology 35,
495-516.
Estall, J.L., Kahn, M., Cooper, M.P., Fisher, F.M., Wu, M.K., Laznik, D., Qu, L., Cohen, D.E.,
Shulman, G.I., Spiegelman, B.M. (2009). Sensitivity of lipid metabolism and insulin signaling to
genetic alterations in hepatic peroxisome proliferator-activated receptor-gamma coactivator-
1alpha expression. Diabetes 58, 1499-1508.
97
Evans, M.J., and Scarpulla, R.C. (1990) NRF-1: a trans-activator of nuclear-encoded respirator
genes in animal cells. Genes Dev 4, 1023-1034.
Fain, J.N. (2006). Release of interleukins and other inflammatory cytokines by human adipose
tissue is enhanced in obesity and primarily due to the nonfat cells. Vitamins and hormones 74,
443-477.
Fan, M., Rhee, J., St-Pierre, J., Handschin, C., Puigserver, P., Lin, J., Jaeger, S., Erdjument-
Bromage, H., Tempst, P., and Spiegelman, B.M. (2004). Suppression of mitochondrial
respiration through recruitment of p160 myb binding protein to PGC-1alpha: modulation by p38
MAPK. Genes & development 18, 278-289.
Fernandez-Marcos, P.J., and Auwerx, J. (2011). Regulation of PGC-1alpha, a nodal regulator of
mitochondrial biogenesis. The American journal of clinical nutrition 93, 884s-890.
Fineman, M.S., Bicsak, T.A., Shen, L.Z., Taylor, K., Gaines, E., Varns, A., Kim, D., and Baron,
A.D. (2003). Effect on glycemic control of exenatide (synthetic exendin-4) additive to existing
metformin and/or sulfonylurea treatment in patients with type 2 diabetes. Diabetes care 26, 2370-
2377.
Fisher, F.M., Kleiner, S., Douris, D., Fox, E.C., Mepani, R.J., Verdeguer, F., Wu, J.,
Kharitonenkov, A., Flier, J.S., Maratos-Flier, E., et al. (2012). FGF21 regulates PGC-1α and
browning of white adipose tissues in adaptive thermogenesis. Genes Dev 26, 271-281.
Forbes, J.M., and Cooper, M.E. (2013). Mechanisms of diabetic complications. Physiological
reviews 93, 137-188.
Fujita, J., Crane, A.M., Souza, M.K., Dejosez, M., Kyba, M., Flavell, R.A., Thomson, J.A., and
Zwaka, T.P. (2008). Caspase activity mediates the differentiation of embryonic stem cells. Cell
stem cell 2, 595-601.
Garrido, C., Galluzzi, L., Brunet, M., Puig, P. E., Didelot, C., and Kroemer, G. (2006).
Mechanisms of cytochrome c release from mitochondria. Cell Death Differ 13, 1423-1433.
Gavrieli, Y., Sherman, Y., and Ben-Sasson, S.A. (1992). Identification of programmed cell death
in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 119, 493-501.
Ghofaili, K.A., Fung, M., Ao, Z., Meloche, M., Shapiro, R.J., Warnock, G.L., Elahi, D.,
Meneilly, G.S., and Thompson, D.M. (2007). Effect of exenatide on beta cell function after islet
transplantation in type 1 diabetes. Transplantation 83, 24-28.
Handschin, C., Lin, J., Rhee, J., Peyer, A.K., Chin, S., Wu, P.H., Meyer, U.A., and Spiegelman,
B.M. (2005). Nutritional regulation of hepatic heme biosynthesis and porphyria through PGC-
1alpha. Cell 122, 505-515.
Handschin, C., and Spiegelman, B.M. (2006). Peroxisome proliferator-activated receptor gamma
coactivator 1 coactivators, energy homeostasis, and metabolism. Endocrine reviews 27, 728-735.
98
Hara, K., Tobe, K., Okada, T., Kadowaki, H., Akanuma, Y., Ito, C., Kimura, S., and Kadowaki,
T. (2002). A genetic variation in the PGC-1 gene could confer insulin resistance and
susceptibility to Type II diabetes. Diabetologia 45, 740-743.
Hotamisligil, G.S., Shargill, N.S., and Spiegelman, B.M. (1993). Adipose expression of tumor
necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science (New York, N.Y.)
259, 87-91.
Huang, S., Rutkowsky, J.M., Snodgrass, R.G., Ono-Moore, K.D., Schneider, D.A., Newman,
J.W., Adams, S.H., and Hwang, D.H. (2012). Saturated fatty acids activate TLR-mediated
proinflammatory signaling pathways. Journal of lipid research 53, 2002-2013.
IDF (2013). Healthcare Expenditures.
Inada, A., Nienaber, C., Katsuta, H., Fujitani, Y., Levine, J., Morita, R., Sharma, A., and Bonner-
Weir, S. (2008). Carbonic anhydrase II-positive pancreatic cells are progenitors for both
endocrine and exocrine pancreas after birth. Proceedings of the National Academy of Sciences of
the United States of America 105, 19915-19919.
Jambal, P., Masterson, S., Nesterova, A., Bouchard, R., Bergman, B., Hutton, J.C., Boxer, L.M.,
Reusch, J.E., and Pugazhenthi, S. (2003). Cytokine-mediated down-regulation of the
transcription factor cAMP-response element-binding protein in pancreatic beta-cells. The Journal
of biological chemistry 278, 23055-23065.
Jhala, U.S., Canettieri, G., Screaton, R.A., Kulkarni, R.N., Krajewski, S., Reed, J., Walker, J.,
Lin, X., White, M., and Montminy, M. (2003). cAMP promotes pancreatic beta-cell survival via
CREB-mediated induction of IRS2. Genes & development 17, 1575-1580.
Karlsen, A.E., Hagopian, W.A., Petersen, J.S., Boel, E., Dyrberg, T., Grubin, C.E., Michelsen,
B.K., Madsen, O.D., and Lernmark, A. (1992). Recombinant glutamic acid decarboxylase
(representing the single isoform expressed in human islets) detects IDDM-associated 64,000-
M(r) autoantibodies. Diabetes 41, 1355-1359.
Kim, A., Miller, K., Jo, J., Kilimnik, G., Wojcik, P., and Hara, M. Islet architecture: A
comparative study. Islets 1, 129-136.
Kim, J.W., You, Y.H., Ham, D.S., Cho, J.H., Ko, S.H., Song, K.H., Son, H.Y., Suh-Kim, H.,
Lee, I.K., and Yoon, K.H. (2009). Suppression of peroxisome proliferator-activated receptor
gamma-coactivator-1alpha normalizes the glucolipotoxicity-induced decreased BETA2/NeuroD
gene transcription and improved glucose tolerance in diabetic rats. Endocrinology 150, 4074-
4083.
Lacquemant, C., Chikri, M., Boutin, P., Samson, C., and Froguel, P. (2002). No association
between the G482S polymorphism of the proliferator-activated receptor-gamma coactivator-1
(PGC-1) gene and Type II diabetes in French Caucasians. Diabetologia 45, 602-603; author reply
604.
99
Lee, J.H., Jo, J., Hardikar, A.A., Periwal, V., and Rane, S.G. (2010). Cdk4 regulates recruitment
of quiescent beta-cells and ductal epithelial progenitors to reconstitute beta-cell mass. PloS one
5, e8653.
Lee, J.Y., Ristow, M., Lin, X., White, M.F., Magnuson, M.A., and Hennighausen, L. (2006).
RIP-Cre revisited, evidence for impairments of pancreatic beta-cell function. J Biol Chem 281,
2649-2653.
Lehman, J.J., Barger, P.M., Kovacs, A., Saffitz, J.E., Medeiros, D.M., and Kelly, D.P. (2000)
Peroxisome proliferator-activated receptor gamma coactivator-1 promotes cardiac mitochondrial
biogenesis. J Clin Invest 106, 847-856.
Leiter, E.H., Prochazka, M., and Coleman, D.L. (1987). The non-obese diabetic (NOD) mouse.
The American journal of pathology 128, 380-383.
Leone, T.C., Lehman, J.J., Finck, B.N., Schaeffer, P.J., Wende, A.R., Boudina, S., Courtois, M.,
Wozniak, D.F., Sambandam, N., Bernal-Mizrachi, C., et al. (2005). PGC-1alpha deficiency
causes multi-system energy metabolic derangements: muscle dysfunction, abnormal weight
control and hepatic steatosis. PLoS Biol 3, e101.
Li, L., El-Kholy, W., Rhodes, C.J., and Brubaker, P.L. (2005). Glucagon-like peptide-1 protects
beta cells from cytokine-induced apoptosis and necrosis: role of protein kinase B. Diabetologia
48, 1339-1349.
Li, Y., Hansotia, T., Yusta, B., Ris, F., Halban, P.A., and Drucker, D.J. (2003). Glucagon-like
peptide-1 receptor signaling modulates beta cell apoptosis. The Journal of biological chemistry
278, 471-478.
Li, Z., Karlsson, F.A., and Sandler, S. (2000). Islet loss and alpha cell expansion in type 1
diabetes induced by multiple low-dose streptozotocin administration in mice. The Journal of
endocrinology 165, 93-99.
Liang, H., Balas, B., Tantiwong, P., Dube, J., Goodpaster, B.H., O’Doherty, R.M., DeFronzo,
R.A., Richardson, A., Musi, N., and Ward, W.F. (2009). Whole body overexpression of PGC-
1alpha has opposite effects on hepatic and muscle insulin sensitivity. Am J Physiol Endocrinol
Metab 296, 945-954.
Lin, J., Handschin, C., and Spiegelman, B.M. (2005). Metabolic control through the PGC-1
family of transcription coactivators. Cell Metab 1, 361-370.
Lin, J., Puigserver, P., Donovan, J., Tarr, P., and Spiegelman, B.M. (2002a). Peroxisome
proliferator-activated receptor gamma coactivator 1beta (PGC-1beta ), a novel PGC-1-related
transcription coactivator associated with host cell factor. The Journal of biological chemistry
277, 1645-1648.
100
Lin, J., Wu, P.H., Tarr, P.T., Lindenberg, K.S., St-Pierre, J., Zhang, C.Y., Mootha, V.K., Jager,
S., Vianna, C.R., Reznick, R.M., et al. (2004). Defects in adaptive energy metabolism with CNS-
linked hyperactivity in PGC-1alpha null mice. Cell 119, 121-135.
Lin, J., Wu, H., Tarr, P.T., Zhang, C.Y., Wu, Z., Boss, O., Michael, L.F., Puigserver, P., Isotani,
E., Olson, E.N., et al. (2002b). Transcriptional co-activator PGC-1 alpha drives the formation of
slow-twitch muscle fibres. Nature 418, 797-801.
Lin, Y.C., Chang, P.F., Chang, M.H., and Ni, Y.H. (2013) A common variant in the peroxisome
proliferator-activated receptor-γ coactivator-1α gene is associated with non-alcoholic fatty liver
disease in obese children. Am J Clin Nutr 97, 326-331.
Ling, C., Del Guerra, S., Lupi, R., Ronn, T., Granhall, C., Luthman, H., Masiello, P., Marchetti,
P., Groop, L., and Del Prato, S. (2008). Epigenetic regulation of PPARGC1A in human type 2
diabetic islets and effect on insulin secretion. Diabetologia 51, 615-622.
Locksley, R.M., Killeen, N., and Lenardo, M.J. (2001). The TNF and TNF receptor
superfamilies: integrating mammalian biology. Cell 104, 487-501.
Lovshin, J.A., and Drucker, D.J. (2009). Incretin-based therapies for type 2 diabetes mellitus.
Nature reviews. Endocrinology 5, 262-269.
Lowell, B.B., and Shulman, G.I. (2005). Mitochondrial dysfunction and type 2 diabetes. Science
(New York, N.Y.) 307, 384-387.
Maechler, P., Li, N., Casimir, M., Vetterli, L., Frigerio, F., and Brun, T. (2010). Role of
mitochondrial in beta-cell function and dysfunction. Adv Exp Med Biol 654, 193-216.
Maestre, I., Jordan, J., Calvo, S., Reig, J.A., Cena, V., Soria, B., Prentki, M., and Roche, E.
(2003). Mitochondrial dysfunction is involved in apoptosis induced by serum withdrawal and
fatty acids in the beta-cell line INS-1. Endocrinology 144, 335-345.
Mathers, C.D., and Loncar, D. (2006). Projections of global mortality and burden of disease from
2002 to 2030. PLoS medicine 3, e442.
McComb, S., Mulligan, R., and Sad, S. (2010). Caspase-3 is transiently activated without cell
death during early antigen driven expansion of CD8(+) T cells in vivo. PloS one 5, e15328.
Meier, C.A., Bobbioni, E., Gabay, C., Assimacopoulos-Jeannet, F., Golay, A., and Dayer, J.M.
(2002). IL-1 receptor antagonist serum levels are increased in human obesity: a possible link to
the resistance to leptin? The Journal of clinical endocrinology and metabolism 87, 1184-1188.
Millay, D.P., and Olson, E.N. (2013). Making muscle or mitochondria by selective splicing of
PGC-1alpha. Cell metabolism 17, 3-4.
101
Miura, S., Kai, Y., Kamei, Y., and Ezaki, O. (2008). Isoform-specific increases in murine
skeletal muscle peroxisome proliferator-activated receptor-gamma coactivator-1alpha (PGC-
1alpha) mRNA in response to beta2-adrenergic receptor activation and exercise. Endocrinology
149, 4527-4533.
Monti, P., Scirpoli, M., Maffi, P., Ghidoli, N., De Taddeo, F., Bertuzzi, F., Piemonti, L., Falcone,
M., Secchi, A., and Bonifacio, E. (2008). Islet transplantation in patients with autoimmune
diabetes induces homeostatic cytokines that expand autoreactive memory T cells. The Journal of
clinical investigation 118, 1806-1814.
Mootha, V.K., Handschin, C., Arlow, D., Xie, X., St Pierre, J., Sihag, S., Yang, W., Altshuler,
D., Puigserver, P., Patterson, N., et al. (2004). Erralpha and Gabpa/b specify PGC-1alpha-
dependent oxidative phosphorylation gene expression that is altered in diabetic muscle.
Proceedings of the National Academy of Sciences of the United States of America 101, 6570-
6575.
Mootha, V.K., Lindgren, C.M., Eriksson, K.F., Subramanian, A., Sihag, S., Lehar, J., Puigserver,
P., Carlsson, E., Ridderstrale, M., Laurila, E., et al. (2003). PGC-1alpha-responsive genes
involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nature
genetics 34, 267-273.
Morgan, N.G., Cable, H.C., Newcombe, N.R., and Williams, G.T. (1994). Treatment of cultured
pancreatic B-cells with streptozotocin induces cell death by apoptosis. Bioscience reports 14,
243-250.
Morrish, N.J., Wang, S.L., Stevens, L.K., Fuller, J.H., and Keen, H. (2001). Mortality and causes
of death in the WHO Multinational Study of Vascular Disease in Diabetes. Diabetologia 44
Suppl 2, S14-21.
Murea, M., Ma, L., and Freedman, B.I. (2012). Genetic and environmental factors associated
with type 2 diabetes and diabetic vascular complications. The review of diabetic studies : RDS 9,
6-22.
Nauck, M.A., Heimesaat, M.M., Orskov, C., Holst, J.J., Ebert, R., and Creutzfeldt, W. (1993).
Preserved incretin activity of glucagon-like peptide 1 [7-36 amide] but not of synthetic human
gastric inhibitory polypeptide in patients with type-2 diabetes mellitus. The Journal of clinical
investigation 91, 301-307.
Nguyen, M.T., Favelyukis, S., Nguyen, A.K., Reichart, D., Scott, P.A., Jenn, A., Liu-Bryan, R.,
Glass, C.K., Neels, J.G., and Olefsky, J.M. (2007). A subpopulation of macrophages infiltrates
hypertrophic adipose tissue and is activated by free fatty acids via Toll-like receptors 2 and 4 and
JNK-dependent pathways. The Journal of biological chemistry 282, 35279-35292.
Norrbom, J., Sallstedt, E.K., Fischer, H., Sundberg, C.J., Rundqvist, H., and Gustafsson, T.
(2011). Alternative splice variant PGC-1alpha-b is strongly induced by exercise in human
skeletal muscle. American journal of physiology. Endocrinology and metabolism 301, E1092-
1098.
102
O'Brien, B.A., Harmon, B.V., Cameron, D.P., and Allan, D.J. (1996). Beta-cell apoptosis is
responsible for the development of IDDM in the multiple low-dose streptozotocin model. The
Journal of pathology 178, 176-181.
Oberkofler, H., Linnemayr, V., Weitgasser, R., Klein, K., Xie, M., Iglseder, B., Krempler, F.,
Paulweber, B., and Patsch, W. (2004). Complex haplotypes of the PGC-1alpha gene are
associated with carbohydrate metabolism and type 2 diabetes. Diabetes 53, 1385-1393.
Olsson, A.H., Yang, B.T., Hall, E., Taneera, J., Salehi, A., Nitert, M.D., and Ling, C. (2011).
Decreased expression of genes involved in oxidative phosphorylation in human pancreatic islets
from patients with type 2 diabetes. European journal of endocrinology / European Federation of
Endocrine Societies 165, 589-595.
Padgett, L.E., Broniowska, K.A., Hansen, P.A., Corbett, J.A., and Tse, H.M. (2013). The role of
reactive oxygen species and proinflammatory cytokines in type 1 diabetes pathogenesis. Annals
of the New York Academy of Sciences 1281, 16-35.
Palomer, X., Alvarez-Guardia, D., Rodriguez-Calvo, R., Coll, T., Laguna, J.C., Davidson, M.M.,
Chan, T.O., Feldman, A.M., and Vazquez-Carrera, M. (2009). TNF-alpha reduces PGC-1alpha
expression through NF-kappaB and p38 MAPK leading to increased glucose oxidation in a
human cardiac cell model. Cardiovasc Res 81, 703-712.
Prentki, M., and Nolan, C.J. (2006). Islet beta cell failure in type 2 diabetes. The Journal of
clinical investigation 116, 1802-1812.
Puigserver, P., Adelmant, G., Wu, Z., Fan, M., Xu, J., O'Malley, B., and Spiegelman, B.M.
(1999). Activation of PPARgamma coactivator-1 through transcription factor docking. Science
(New York, N.Y.) 286, 1368-1371.
Puigserver, P., Rhee, J., Donovan, J., Walkey, C.J., Yoon, J.C., Oriente, F., Kitamura, Y.,
Altomonte, J., Dong, H., Accili, D., et al. (2003). Insulin-regulated hepatic gluconeogenesis
through FOXO1-PGC-1alpha interaction. Nature 423, 550-555.
Puigserver, P., Rhee, J., Lin, J., Wu, Z., Yoon, J.C., Zhang, C.Y., Krauss, S., Mootha, V.K.,
Lowell, B.B., and Spiegelman, B.M. (2001) Cytokine stimulation of energy expenditure through
p38 MAP kinase activation of PPARgamma coactivator-1. Mol Cell 8, 971-982.
Puigserver, P., Wu, Z., Park, C.W., Graves, R., Wright, M., and Spiegelman, B.M. (1998). A
cold-inducible coactivator of nuclear receptors linked to adaptive thermogenesis. Cell 92, 829-
839.
Rehman, K.K., Trucco, M., Wang, Z., Xiao, X., and Robbins, P.D. (2008). AAV8-mediated gene
transfer of interleukin-4 to endogenous beta-cells prevents the onset of diabetes in NOD mice.
Molecular therapy : the journal of the American Society of Gene Therapy 16, 1409-1416.
Rehman, K.K., Wang, Z., Bottino, R., Balamurugan, A.N., Trucco, M., Li, J., Xiao, X., and
Robbins, P.D. (2005). Efficient gene delivery to human and rodent islets with double-stranded
(ds) AAV-based vectors. Gene therapy 12, 1313-1323.
103
Reusch, J.E., and Wang, C.C. (2011). Cardiovascular disease in diabetes: where does glucose fit
in? The Journal of clinical endocrinology and metabolism 96, 2367-2376.
Rhee, J., Inoue, Y., Yoon, J.C., Puigserver, P., Fan, M., Gonzalez, F.J., and Spiegelman, B.M.
(2003). Regulation of hepatic fasting response by PPARgamma coactivator-1alpha (PGC-1):
requirement for hepatocyte nuclear factor 4alpha in gluconeogenesis. Proceedings of the
National Academy of Sciences of the United States of America 100, 4012-4017.
Riedel, M.J., Gaddy, D.F., Asadi, A., Robbins, P.D., and Kieffer, T.J. (2010). DsAAV8-
mediated expression of glucagon-like peptide-1 in pancreatic beta-cells ameliorates
streptozotocin-induced diabetes. Gene therapy 17, 171-180.
Ruas, J.L., White, J.P., Rao, R.R., Kleiner, S., Brannan, K.T., Harrison, B.C., Greene, N.P., Wu,
J., Estall, J.L., Irving, B.A., et al. (2012). A PGC-1alpha isoform induced by resistance training
regulates skeletal muscle hypertrophy. Cell 151, 1319-1331.
Sakata, N., Yoshimatsu, G., Tsuchiya, H., Egawa, S., and Unno, M. (2012). Animal models of
diabetes mellitus for islet transplantation. Experimental diabetes research 2012, 256707.
Sano, M., Tokudome, S., Shimizu, N., Yoshikawa, N., Ogawa, C., Shirakawa, K., Endo, J.,
Katayama, T., Yuasa, S., Ieda, M., et al. (2007). Intramolecular control of protein stability,
subnuclear compartmentalization, and coactivator function of peroxisome proliferator-activated
receptor gamma coactivator 1alpha. The Journal of biological chemistry 282, 25970-25980.
Schiff, M., Loublier S., Coulibaly, A., Benit, P., de Baulny, H.O., and Rustin, P. (2009).
Mitochondria and diabetes mellitus: untangling a conflictive relationship? J Inherit Metab Dis
32, 684-698.
Schnedl, W.J., Ferber, S., Johnson, J.H., and Newgard, C.B. (1994). STZ transport and
cytotoxicity. Specific enhancement in GLUT2-expressing cells. Diabetes 43, 1326-1333.
Schuler, M., Ali, F., Chambon, C., Duteil, D., Bornert, J.M., Tardivel, A., Desvergne, B., Wahli,
W., Chambon, P., and Metzger, D. (2006). PGC1alpha expression is controlled in skeletal
muscles by PPARbeta, whose ablation results in fiber-type switching, obesity and type 2
diabetes. Cell Metabolism 4, 407-414.
Schon, E.A., and Manfredi, G. (2003). Neuronal degeneration and mitochondrial dysfunction.
The Journal of clinical investigation 111, 303-312.
Schreiber, S.N., Emter, R., Hock, M.B., Knutti, D., Cardenas, J., Podvinec, M., Oakelet, E.J., and
Kralli, A. (2004). The estrogen-related receptor alpha (ERRalpha) functions in PPARgamma
coactivator 1alpha (PGC-1alpha)-induced mitochondrial biogenesis. Proc Natl Acad Sci 101,
6472-6477.
Shapiro, A.M., Lakey, J.R., Ryan, E.A., Korbutt, G.S., Toth, E., Warnock, G.L., Kneteman,
N.M., and Rajotte, R.V. (2000). Islet transplantation in seven patients with type 1 diabetes
mellitus using a glucocorticoid-free immunosuppressive regimen. The New England journal of
medicine 343, 230-238.
104
Sivitz, W.I., and Yorek, M.A. (2010). Mitochondrial dysfunction in diabetes: from molecular
mechanisms to functional significance and therapeutic opportunities. Antioxid Redx Signal 12,
537-577.
Steil, G.M., Trivedi, N., Jonas, J.C., Hasenkamp, W.M., Sharma, A., Bonner-Weir, S., and Weir,
G.C. (2001). Adaptation of beta-cell mass to substrate oversupply: enhanced function with
normal gene expression. American journal of physiology. Endocrinology and metabolism 280,
E788-796.
Szkudelski, T. (2001). The mechanism of alloxan and streptozotocin action in B cells of the rat
pancreas. Physiological research / Academia Scientiarum Bohemoslovaca 50, 537-546.
Valtat, B., Riveline, J.P., Zhang, P., Singh-Estivalet, A., Armanet, M., Venteclef, N., Besseiche,
A., Kelly, D.P., Tronche, F., Ferre, P., et al. (2013). Fetal PGC-1alpha overexpression programs
adult pancreatic beta-cell dysfunction. Diabetes 62, 1206-1216.
van Belle, T.L., Coppieters, K.T., and von Herrath, M.G. (2011). Type 1 diabetes: etiology,
immunology, and therapeutic strategies. Physiological reviews 91, 79-118.
Vega, R.B., Huss, J.M., and Kelly, D.P. (2000). The coactivator PGC-1 cooperates with
peroxisome proliferator-activated receptor alpha in transcriptional control of nuclear genes
encoding mitochondrial fatty acid oxidation enzymes. Molecular and cellular biology 20, 1868-
1876.
Virbasius, J.V., Virbasius, C.A., and Scarpulla, R.C. Identity of GABP with NRF-2, a
multisubunit activator of cytochrome oxidase expression, reveals a cellular role for an ETS
domain activator of viral promoters. Genes Dev 7, 380-392.
Wallberg, A.E., Yamamura, S., Malik, S., Spiegelman, B.M., and Roeder, R.G. (2003).
Coordination of p300-mediated chromatin remodeling and TRAP/mediator function through
coactivator PGC-1alpha. Molecular cell 12, 1137-1149.
Wang, Q., and Brubaker, P.L. (2002). Glucagon-like peptide-1 treatment delays the onset of
diabetes in 8 week-old db/db mice. Diabetologia 45, 1263-1273.
Wang, Z., Zhu, T., Rehman, K.K., Bertera, S., Zhang, J., Chen, C., Papworth, G., Watkins, S.,
Trucco, M., Robbins, P.D., et al. (2006). Widespread and stable pancreatic gene transfer by
adeno-associated virus vectors via different routes. Diabetes 55, 875-884.
Weisberg, S.P., McCann, D., Desai, M., Rosenbaum, M., Leibel, R.L., and Ferrante, A.W., Jr.
(2003). Obesity is associated with macrophage accumulation in adipose tissue. The Journal of
clinical investigation 112, 1796-1808.
Wenz, T., Rossi, S.G., Rotundo, R.L., Spiegelman, B.M., and Moraes, C.T. (2009). Increased
muscle PGC-1alpha expression protects from sarcopenia and metabolic disease during aging.
Proceedings of the National Academy of Sciences of the United States of America 106, 20405-
20410.
105
Westerbacka, J., Kolak, M., Kiviluoto, T., Arkkila, P., Siren, J., Hamsten, A., Fisher, R.M., and
Yki-Jarvinen, H. (2007). Genes involved in fatty acid partitioning and binding, lipolysis,
monocyte/macrophage recruitment, and inflammation are overexpressed in the human fatty liver
of insulin-resistant subjects. Diabetes 56, 2759-2765.
Woo, M., Hakem, R., Furlonger, C., Hakem, A., Duncan, G.S., Sasaki, T., Bouchard, D., Lu, L.,
Wu, G.E., Paige, C.J., et al. (2003). Caspase-3 regulates cell cycle in B cells: a consequence of
substrate specificity. Nature immunology 4, 1016-1022.
Wu, K.K., and Huan, Y. (2008). Streptozotocin-induced diabetic models in mice and rats.
Current protocols in pharmacology / editorial board, S.J. Enna (editor-in-chief) ... [et al.] Chapter
5, Unit 5.47.
Wu, Z., Puigserver, P., Andersson, U., Zhang, C., Adelmant, G., Mootha, V., Troy, A., Cinti, S.,
Lowell, B., Scarpulla, R.C., et al. (1999). Mechanisms controlling mitochondrial biogenesis and
respiration through the thermogenic coactivator PGC-1. Cell 98, 115-124.
Xu, H., Barnes, G. T., Yang, Q., Tan, G., Yang, D., Chou, C.J., Sole, J., Nichols, A., Ross, J.S.,
Tartaglia, L, A., et al. (2003). Chronic inflammation in fat plays a crucial role in the development
of obesity-related insulin resistance. J Clin Invest 112, 1821-1830
Yoon, J.C., Xu, G., Deeney, J.T., Yang, S.N., Rhee, J., Puigserver, P., Levens, A.R., Yang, R.,
Zhang, C.Y., Lowell, B.B., et al. (2003). Suppression of beta cell energy metabolism and insulin
release by PGC-1alpha. Developmental cell 5, 73-83.
Young, A.A., Gedulin, B.R., Bhavsar, S., Bodkin, N., Jodka, C., Hansen, B., and Denaro, M.
(1999). Glucose-lowering and insulin-sensitizing actions of exendin-4: studies in obese diabetic
(ob/ob, db/db) mice, diabetic fatty Zucker rats, and diabetic rhesus monkeys (Macaca mulatta).
Diabetes 48, 1026-1034.
Yusta, B., Baggio, L.L., Estall, J.L., Koehler, J.A., Holland, D.P., Li, H., Pipeleers, D., Ling, Z.,
and Drucker, D.J. (2006). GLP-1 receptor activation improves beta cell function and survival
following induction of endoplasmic reticulum stress. Cell Metab 4, 391-406.
Zhang, P., Liu, C., Zhang, C., Zhang, Y., Shen, P., Zhang, J., and Zhang, C.Y. (2005). Free fatty
acids increase PGC-1alpha expression in isolated islets. FEBS Letters 579, 1446-1456.
Zhang, Y., Huypens, P., Adamson, A.W., Chang, J.S., Henagan, T.M., Boudreau, A., Lenard,
N.R., Burk, D., Klein, J., Perwitz, N., et al. (2009). Alternative mRNA splicing produces a novel
biologically active short isoform of PGC-1alpha. The Journal of biological chemistry 284,
32813-32826.
Zhang, Y., Zhang, Y., Bone, R.N., Cui, W., Peng, J.B., Siegal, G.P., Wang, H., and Wu, H.
(2012). Regeneration of pancreatic non-beta endocrine cells in adult mice following a single
diabetes-inducing dose of streptozotocin. PloS one 7, e36675.
Recommended